ARTICLE

TRUTHS IN TRANSLATION

Frédéric Gilles Sourgens*

I. INTRODUCTION ...... 102 II. A PRIMA FACIE LACK OF FACTUAL RIGOR ...... 107 A. Burden of Proof ...... 111 1. The Basic Rule in Action ...... 111 2. Presumptions Shift the Basic Rule ...... 115 3. Burdens, Presumptions, and (Post-)Truth ...... 117 B. Standard of Proof ...... 118 C. Burdens and Standards Cannot Operate Without Presumptions ...... 124 III. INFERENCES AND PRESUMPTIONS ...... 124 A. Proof by Direct Evidence ...... 124 1. Contemporaneous Documentary Evidence...... 126 2. Witness Testimony ...... 132 3. The Quest for “Corroboration” ...... 134 B. Inferences ...... 138 1. Inferences as Gap Fillers ...... 139 2. The Threshold Question of Reasonableness ..... 141 3. Plausibility—The Choice Between Reasonable Alternatives ...... 142 C. Presumptions ...... 146 1. Legal Presumptions as Part of Applicable Law 147

* Senator Robert J. Dole Distinguished Professor of Law & Director, Oil and Gas Law Center, Washburn University School of Law; Editor in Chief, InvestmentClaims.com; former Co-Chair, American Society of International Law Private International Law Interest Group. I would like to thank the participants in the Oxford University Press Investment Claims Summer Academy in Oxford, UK, as well the participants in the First Global Land and Resources Symposium held in Norman, OK for their insightful comments. I would also like to thank Kelsey Halloran and the exceptional student editors and staff working on Volume 44 of the Fordham International Law Journal. Their editorial assistance has been invaluable.

101 102 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1

2. Judicial Presumptions Based on Relevant Practice ...... 148 IV. PRESUMPTIONS’ BIAS PROBLEM ...... 152 A. The Theoretical Foundation of Presumptions ...... 154 B. Presumptions as Narrative ...... 157 C. The Bias Problem – How to Test Narratives? ...... 162 V. PRESUMPTIVE POLYVALENCE ...... 165 A. Presumptions as Translation ...... 165 B. Connotations and Presumptions ...... 176 C. Deploying Translating Presumptions ...... 179 VI. CONCLUSION ...... 185

“‘Thee it behoves to take another road.’” (Virgil to Dante) The Divine Comedy, Inferno, Canto I1

I. INTRODUCTION It has sadly become trite to observe that we live in a post-truth society.2 The hallmark of a post-truth society is its reliance upon alternative facts.3 Such alternative facts are no longer grounded in empirical evidence or governed by the basic laws of logic.4 Instead,

1. DANTE ALIGHIERI, THE DIVINE COMEDY 9 (Henry Wadsworth Longfellow, trans., Barnes & Nobles Inc., 2016). This article is about the value of stepping off a seemingly direct path and finding a different way to one’s goal that is premised in a broader quest for understanding. The conversation between Dante and Virgil is one such stepping off— Dante found his way barred and looked for a guide to take another road. This road is allegorical in the Divine Comedy. But importantly for this article, it is also programmatic for the act of writing the poem: Dante translates Virgil, Latin to Italian, epic hexameter to terza rima, and the adoration of Venus and Caesar to a Christian context. Longfellow in turn translates Dante—Italian into English, European high middle ages to American romanticism and classicism to an opening to modern literatures. Our understanding of truth and identity, and our exercise of judgment changes when we take that different road. And in taking it, more often than not, translation is a key map to keep us from getting lost in the woods. 2. See S.I. Strong, Alternative Facts and the Post-Truth Society: Meeting the Challenge, 165 U. PA. L. REV. ONLINE 137, 137-38 (2017) (discussing the post-truth problem). 3. Id. 4. Compare Trump Holds Firm for Kavanaugh but Calls Accuser Credible, ASSOC. PRESS (Sept. 28, 2018), https://apnews.com/2650e44f2a73484da4c29ca4343dc6f5 [https://perma.cc/22M5-NCZS] (President Trump finding testimony by a woman accusing Brett Kavanaugh of sexual assault “very compelling”), with Peter Baker, Trump Bets 2020] TRUTHS IN TRANSLATION 103 commentators submit that public discourse today is dominated by those who peddle in narrative with callous disregard for the truth.5 Lawyers frequently assert that the rule of law is a potent antidote against such an attitude to the truth.6 The rule of law was thought of as a check against tin-pot dictators and authoritarian rule after the fall of the .7 The link is certainly cemented deeply enough within our collective subconscious through the clockwork repetition of images of young dashing lawyers (frequently played by Tom Cruise) extracting a truth we supposedly cannot handle from mendacious authority figures in the name of law and justice.8 The problem is that—on the world stage at least—this assertion may be a myth. If one were to ask an international judge how she would get to the truth or examine the evidence, she would tell us “I am guided by my intimate conviction.”9 While her intimate conviction might differ from that of, say, President Vladimir Putin of Russia or Senator Lindsey Graham of South Carolina with regard to a specific brewing international incident, she could not in fact

Kavanaugh ‘Hoax’ Turns into Midterm Gains, N.Y. TIMES (Oct. 8, 2018), https://www.nytimes.com/2018/10/08/us/politics/trump-kavanaugh-accusations- hoax.html [https://perma.cc/3UNY-STGP] (President Trump asserting the same accusation is “‘a hoax’ and ‘fabricated.’”). 5. See Quinta Jurecic, On Bullshit and the Oath of Office: The “LOL Nothing Matters” Presidency, LAWFARE (Nov. 23, 2016, 11:28 AM), https://www.lawfareblog.com/bullshit- and-oath-office-lol-nothing-matters-presidency [https://perma.cc/B5QU-P8AS]; Roger Cohen, Donald Trump Just Cannot Help It, N.Y. TIMES (Jan. 11, 2019), https://www.nytimes.com/2019/01/11/opinion/donald-trump-illegal-immigration- border-wall.html [https://perma.cc/8YF9-F3XQ]; HARRY G. FRANKFURT, ON BULLSHIT passim (2005). 6. See Thomas Burri, Regulating the Risk of Trumpism, 8 EUR. J. RISK REG. 64, 64-65 (2017) (“New ways need to be found to fortify the rule of law against lies, hatred, and violence which, in a post-truth age, spread unfiltered”); but see David S. Rubenstein, Taking Care of the Rule of Law, 86 GEO. WASH. L. REV. 168, 205 (2018) (“Especially in a ‘post-truth’ era, where perceptions dominate over facts, rule of law arguments that turn on contestable facts and framings will never convince non-believers.”). 7. John K. M. Ohnesorge, Developing Development Theory: Law and Development Orthodoxies and the Northeast Asian Experience, 28 U. PA. J. INT’L ECON. L. 219, 247-48 (2007). 8. A FEW GOOD MEN (Castle Rock Entertainment 1992) (Kaffee: “I WANT THE TRUTH!” Jessup: “YOU CAN’T HANDLE THE TRUTH!”); see also THE FIRM (Davis Entertainment, Mirage Enterprises 1993) (“It’s not sexy, but it’s got teeth! Ten thousand dollars and five years in prison. That’s ten and five for each act. Have you really looked at that?”). 9. See PHILIPPA WEBB, INTERNATIONAL JUDICIAL INTEGRATION AND FRAGMENTATION 189 (2013) (discussing the use of the intimate conviction standard in international law). 104 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 tell either of them how she approached the evidence differently from both men in advancing their respective political claims.10 All she could have told us is to trust her (and, perhaps, not them). This blind spot is deeply problematic. As this Article submits, the rule of law on the world stage is at risk of being guided by narrative as much as the post-truth populism against which it is invoked. Post-truth populists use narratives to anchor seemingly empirical statements (“largest crowd ever”) in a group’s own mythologies of self-worth (“we are the moral majority”).11 In the legal context, narratives are similarly omni-present. Narratives in the legal context guide factfinding through the central use of presumptions. These presumptions come to be used when there is inconclusive evidence to determine what took place. Centrally, decisionmakers in those instances look at the record through the lens of a presumption that actors conduct themselves in good faith.12 They acted as they ought to have acted. This approach dangerously short circuits the fact-finding process. In more concrete terms, Professors Oona Hathaway and Scott Shapiro have studied an exhaustive historical record of pre-1928 war manifestos.13 They explain that “[b]y making the reasons for war manifest, manifestos sought to make clear that the war in question was just.”14 A just war, in turn, was waged on permissible legal grounds, meaning that the manifestos referred to serious

10. Julia Ioffe, What Putin Really Wants, ATLANTIC (Feb. 2018), https://www.theatlantic.com/magazine/archive/2018/01/putins-game/546548/ [https://perma.cc/2W72-6WAZ] (“Putin and Lavrov were known within the Obama administration for their long tirades, chastising the American president for all the disrespect shown to Russia since 1991”); Michael D. Shear, Furious Lindsey Graham Calls Kavanaugh Hearing ‘the Most Unethical Sham’, N.Y. TIMES (Sept. 27, 2018), https://www.nytimes.com/2018/09/27/us/politics/lindsey-graham-kavanaugh- hearing.html [https://perma.cc/SHL4-22YP]. 11. Eric Bradner, Conway: Trump White House Offered ‘Alternative Facts’ on Crowd Size, CNN (Jan. 23, 2017), https://www.cnn.com/2017/01/22/politics/kellyanne- conway-alternative-facts/index.html [https://perma.cc/LDH2-MHUC]; Alternative Facts' Remark Tops 2017 List of Notable Quotes, FOXNEWS.COM (Dec. 12, 2017), https://www.foxnews.com/us/alternative-facts-remark-tops-2017-list-of-notable- quotes [https://perma.cc/82Z2-SUBZ]. 12. BIN CHENG, GENERAL PRINCIPLES OF LAW AS APPLIED BY INTERNATIONAL COURTS AND TRIBUNALS 305 (2006). 13. OONA A. HATHAWAY & SCOTT J. SHAPIRO, THE INTERNATIONALISTS, HOW A RADICAL PLAN TO OUTLAW WAR REMADE THE WORLD 31-81 (2017). 14. Id. at 40. 2020] TRUTHS IN TRANSLATION 105 grievances that sounded in the then-existing international law.15 Applying the presumptions developed in international law that parties act honesty and reasonably, it would follow that no state ever invaded another purely for reasons of conquest. This conclusion is facially naïve. In fact, Hathaway and Shapiro have debunked it as historically inaccurate.16 The very fact-finding process underlying a rule of law approach to conflict resolution thus would be dangerously blind to the actual facts and construct a reality out of its own, narrative-based “alternative” facts. This Article argues that this state of affairs requires us fundamentally to re-think how factfinding processes must work. It argues that existing global factfinding processes can be redeemed if they do not assume that facts can be constructed out of a single narrative. Instead, it submits that factfinding must actively translate between narratives of the parties affected by the dispute. This reconstruction of the factfinding process relies upon the principle of good faith. This principle requires, in broad contours, that parties act with other regard. This Article applies this principle to the factfinding process itself by requiring that facts are established with other regard, that is with due respect to the respective narratives through which the parties encounter the dispute. This approach allows factfinding premised in the rule of law to make sense of factual claims even when they cannot be empirically tested. It focuses not on what was. Such a focus would be unlikely ever to become testable and as such leads to proof by narrative alone. Rather, it focuses on communicative effects. It can create a context in which conduct is meaningful because of the interaction between the narratives of the affected parties, and thus tests narratives against each other and finds the truth in this interplay between narratives. Again, in the context of examining the causes of war, this is what diligent historians do when they examine the stated causes for armed conflict. Professor Christopher Clark’s explanation of how (rather than why) World War I came about is a masterful display of such research: he notes, consistent with the approach of

15. Id. 16. See id. at 42, 97 (noting that manifestos “matter precisely because they are propaganda” and that “Hugo Grotius was not the great apostle of peace. He was the great apologist for war.”). 106 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 looking to communicative effects, that “the focus on how suggests an alternative approach: a journey through the events that is not driven by the need to draw up a charge sheet against this or that state or individual, but aims to identify decisions that brought war about and to understand the reasoning or emotions behind them.”17 He goes on that this “does not mean excluding questions of responsibility entirely from the discussion—the aim is rather to let the why answers grow, as it were, out of the how answers rather than the other way around.”18 It is thus focus on the between—the interaction of decisions in light of what was known to each of the players when the decision was made and the real time knock on effect of each decision. Another historical masterpiece using a similar approach is David Herbert Donald’s biography of Abraham Lincoln, which seeks to explain to us the basis for his decisions as they were made rather than fall prey to hindsight bias.19 As this Article will show, this new approach is fundamentally consistent with best practices in the international jurisprudence. It does, however, permit a more nuanced legal critique of decisions that should be intuitively problematic. It thus advances the understanding of how international law fact-finding can live up to its rule of law aspirations and provides a new doctrinal toolkit to improve the quality of justice before international courts and tribunals. More broadly, this Article shows how the rule of law can in fact become instrumental in engagement in a post-truth society. It creates a means of engagement with others even when parties rely predominantly on narratives. It therefore showcases how the rule of law is not an antagonist to populist rhetoric. Rather, it shows how the application of the rule of law in fact can overcome the challenges of participating in a post-truth discourse. The rule of law, in other words, can continue to live up to its lofty promise and aspirations. To do so, the rule of law must however, become more yielding and listen rather than engaging in the traditional trope of a Tom Cruise-lookalike shouting at authority at the top of his lungs. This Article is organized in five parts. Part I introduces the apparent methodological sloppiness of international judges and

17. CHRISTOPHER CLARK, THE SLEEPWALKERS: HOW EUROPE WENT TO WAR IN 1914 xxviii (2012) (emphasis in original). 18. Id. (emphasis in original). 19. DAVID HERBERT DONALD, LINCOLN 13 (1995). 2020] TRUTHS IN TRANSLATION 107 arbitrators in their engagement with record facts. Part II will then explain that this apparent methodological sloppiness, thankfully, is only skin deep. International courts and tribunals do have set means by which they scrutinize evidence. This means of analyzing and testing evidence is anchored in burdens of proof and standards of proof. These burdens and standards of proof, in turn, are influenced by the presumptions; courts use these presumptions in attempts to do material justice rather than mechanizing claim resolutions. Part III then sets out the doctrinal rules governing the use of presumptions and inferences proper. It explains how presumptions and inferences come to be used in the complete record context and thus underlines again the vital role presumptions play in international justice. Part IV explains the problem encountered by presumptions due to their grounding in the principle of good faith. Part IV explains that the current use of good faith runs into an is-ought problem and becomes reliant upon narrative as a tool of dispute resolution that it is no longer able to test by empirical means. Part V concludes with a normative proposal how to improve the link between truth and (international) rule of law. This proposal is to take seriously the grounding of presumptions in doctrines of good faith as other regard, and explain how this grounding is doctrinally defensible and helps in overcoming the narrative problem created by the necessary reliance upon presumptions in a rule of law based factfinding paradigm.

II. A PRIMA FACIE LACK OF FACTUAL RIGOR Rule of law has long been held out as an antidote to authoritarian rule. As part of the Washington consensus, the rule of law was thought to hem in the margin of maneuver for dictators and populist leaders.20 Significantly, the rule of law was also thought of a means to confront and debunk authoritarian propaganda and return decision-making and international public discourse to a rational engagement of facts.21 Despite its importance as an antidote to propaganda, rule of law advocates have reasonably neglected to explain how this goal

20. Burri, supra note 6, at 64-65. 21. Ohnesorge, supra note 7, at 247-48. 108 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 should be achieved. The closest set of rules to the broader problem we experience is the law of evidence, that is the law governing how those applying the rule of law themselves gather facts. But the law of evidence remains a doctrinally under-developed field. Consequently, the treatment of evidentiary issues in international decisions, as well as in scholarship discussing these decisions frequently lacks apparent rigor. One such example of a facial lack of evidentiary rigor is the Pulp Mills case between Argentina and Uruguay.22 Argentina alleged that Uruguay had acted in violation of the 1975 Statute of the River Uruguay by building pulp mills on its shores, thereby impermissibly affecting the river’s water quality.23 The Court held that Uruguay had breached its procedural obligations to consult but had not breached its substantive obligations under the 1975 Statute.24 Specifically, Uruguay had failed to pass on appropriate environmental impact assessments as required by the treaty.25 Uruguay had not however violated its obligations not to pollute or change the ecological balance of the river.26 As lawyers are not environmental scientists versed in river ecosystems, the Court required the aide of expertise.27 The parties did not however provide the relevant expertise directly.28 Rather, they hired consultants to appear as co-counsel. By doing so, they shielded the consultants from being examined by the Court or opposing counsel.29 Understandably, the Court was frustrated with this strategic choice as it complicated its fact-finding mission and let the parties know as much in its judgment.30 Nevertheless, the Court pressed on. The Court explained that it “does not find it necessary in order to adjudicate the present case to enter into a general discussion on the relative merits, reliability and authority of the documents and

22. Case Concerning Pulp Mills on the River Uruguay (Arg. v. Urug.), Judgment, 2010 I.C.J. 14 (Apr. 20) [hereinafter Pulp Mills]. 23. Id. at 25. 24. Id. at 106. 25. Id. at 60. 26. Id. at 86. 27. Id. at 72. 28. Pulp Mills, 2010 I.C.J at 72. 29. Id. 30. Id. 2020] TRUTHS IN TRANSLATION 109 studies prepared by the experts and consultants of the Parties.”31 What is more, the Court noted (again with a hint of annoyance) that “despite the volume and complexity of the factual information submitted to it,” it would continue with its mission.32 It explained “in keeping with its practice, the Court will make its own determination of the facts, on the basis of the evidence presented to it, and then it will apply the relevant rules of international law to those facts which it has found to have existed.”33 While one can certainly feel the Court’s pain oozing through these passages, a common law trained lawyer would balk at this manner of discussing evidence. How can a case be decided on the merits that by all accounts requires expert evidence without passing “on the relative merits, reliability and authority of the documents and studies prepared by the experts and consultants of the Parties?”34 How exactly can the Court make “its own determination of facts” without such comments? It would seem to be beyond the Court’s ability (or any lawyer’s ability for that matter) to pass judgment on the impact of discharges on Uruguay water qualities without expert evidence.35 This, however, is precisely what the Court appears to imply it is able to do. The discussion of evidentiary standards in Pulp Mills showcases the overall problem well. The Court is aware of an underlying deficiency in the manner in which a case was brought to it for decision.36 Rather than engage the morass of a record created by the parties head on, however, the Court seeks to elevate itself above the fray;37 One could perhaps picture an imaginary motto hanging in chambers in the Hague—“When they go low, we go high.”38 It then seeks to resolve by clean legal analysis what the parties left unresolved through careful factual proof.39 Problematically, however, this exercise stands to succeed only if

31. Id. 32. Id. at 72. 33. Id. 34. Pulp Mills, 2010 I.C.J. at 73. 35. Id. at 91-99. 36. Id. at 72-73. 37. Id. 38. Quotes from Hillary Clinton’s Convention, N.Y. TIMES (July 30, 2016), https://www.nytimes.com/2016/07/31/opinion/sunday/quotes-from-hillary-clintons- convention.html?searchResultPosition=10 [https://perma.cc/Q6TR-XHLY]. 39. Pulp Mills, 2010 I.C.J. at 91-99. 110 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 the Court is able to make sense of what the parties, to its mind, strategically sought to obfuscate. Unless the Court can bring legal tools to bear to sort out the evidentiary mess, it would seem that it could only make a bad decision. The Court is sadly no exception when it comes to the underdevelopment of the treatment of evidence. International arbitration commentators noted that the only rule of evidence in international arbitration was the discretion of the arbitrator.40 These commentators in fact credit the practice before the International Court of Justice as instructive on questions of evidence before international arbitral tribunals.41 Such comments would invite reasonably similar types of decisions in international arbitration, as well. This potential conclusion is deeply problematic. Any legal analysis is only as solid as the factual foundation upon which it is built. Moreover, factual findings frequently drive the result in international arbitrations or adjudications.42 This would suggest that international dispute resolution has a potentially serious foundation problem as well as a serious legitimacy problem. This problem becomes more pronounced when we look to international law fact-finding as an antidote to the post-truth syndrome in current political discourse. The problem with the post-truth syndrome is that there are no hallmarks for how one should anchor discourse in some form of discernible reality.43 Rather, it is possible for discourse participants to disappear into narratives without much risk to be brought back to earth by means of fact checking.44 The discussion so far has shown that international justice has work to do in order to fare much better— or provide the tools by which the rule of law could address the post-truth syndrome more broadly. So far, there does not appear to be a clear matrix according to which the rule of law could reign in the post-truth syndrome. All it can offer so far is the advice to

40. CHRISTOPH SCHREUER ET AL., THE ICSID CONVENTION: A COMMENTARY 641-45 (2d ed. 2009). 41. Id. 42. Stanimir Alexandrov, Remarks, in 2 INVESTMENT TREATY ARBITRATION AND INTERNATIONAL LAW 195, 205 (Ian Laird & Todd Weiler eds., 2009). 43. Strong, supra note 2, at 145. 44. Id. 2020] TRUTHS IN TRANSLATION 111 trust the discretion of appointed decision-makers. In the current political context, such a request is reasonably naïve. II. Burdens and Standards of Proof The starting point for any account how the international rule of law approaches fact-finding is to understand the role played by burdens and standards of proof in international justice. The concepts of burdens and standards of proof are the first principles that allow for an orderly decision-making process in contested cases. They both rest on the insight that there is a difference between allegation and proof—and that the person doing the alleging is also the person who should do the proving. This Part outlines how that particular insight operates in international justice. The key insight it offers is that burdens of proof and standards of proof are not the sterile, blindly operated scales of courthouse art. Rather, burdens and standards of proof are tethered to evidentiary presumptions, which implicitly embed a narrative about the world within the very operation of international justice.

A. Burden of Proof

1. The Basic Rule in Action The International Court of Justice is a foundational source for key concepts of burdens of proof in international justice. Our Pulp Mills test case reveals the extent to which the Court tends to rely on burdens of proof to resolve disputes.45 As already discussed above, the Pulp Mills case involved reasonably complicated scientific questions which were left unaddressed by competent expert testimony and cross-examination.46 The Court therefore appeared to suggest that it would make its own evidentiary findings without going into a detailed examination of the expert submissions.47 A careful analysis of the relevant substantive sections of the Pulp Mills decision leads to a deeply surprising result: the Court did not ultimately end up making factual findings despite its facial promise to do so on the face of the judgment. Argentina advanced

45. Pulp Mills, Judgment, 2010 I.C.J. 14, 14-15 (Apr. 20). 46. Id. at 72-73. 47. Id. 112 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 as its core submissions that Uruguay had failed to (1) contribute to the optimum rational utilization of the river; (2) ensure that the management of the soil and woodlands would not impair the quality of river waters; (3) to coordinate measures to avoid changes in the ecological balance; and (4) to prevent pollution and preserve the aquatic environment by discharging (a) dissolved oxygen, (b) phosphorus, (c) phenolic substances, (d) nonylphenols, (e) dioxins and furans, as well as (f) negatively affecting biodiversity.48 The Court addressed these submissions in turn.49 The Court joined the first submission advanced by Argentina on optimum rational utilization of the river to its substantive analysis of the specific pollution claims advanced by Argentina.50 The Court rejected Argentina’s second submission concerning the management of the soil and woodlands because “Argentina has not provided any evidence to support its contention.”51 It similarly concluded with regard to Argentina’s third contention, i.e., Uruguay had failed to coordinate measures to avoid changes in the ecological balance, that “Argentina has not convincingly demonstrated that Uruguay has refused to engage in such co- ordination as envisaged by Article 36, in breach of that provision.”52 The Court adopted a similar method to address the specific pollution claims raised by Argentina. It ruled as follows with regard to Argentina’s specific claims:

 Dissolved oxygen claim: “the Court finds that the allegation made by Argentina remains unproven.”53

 Phosphorous claim: “The Court finds that based on the evidence before it, the Orion (Botnia) mill has so far complied with the standard for total phosphorus in effluent discharge.”54

48. Id. at 14-55. 49. Id. 50. Id. at 74 (“Of particular relevance in the present case are its functions relating to rule-making in respect of conservation and preservation of living resources, the prevention of pollution and its monitoring, and the co-ordination of actions of the Parties. These functions will be examined by the Court in its analysis of the positions of the Parties with respect to the interpretation and application of Articles 36 and 41 of the 1975 Statute.”). 51. Pulp Mills, 2010 I.C.J. at 75. 52. Id. at 77. 53. Id. at 93. 54. Id. at 95. 2020] TRUTHS IN TRANSLATION 113

 Phenolic substances claim: “[b]ased on the record, and the data presented by the Parties, the Court concludes that there is insufficient evidence to attribute the alleged increase in the level of concentrations of phenolic substances in the river to the operations of the Orion (Botnia) mill.”55

 Nonylphenols claim: “Argentina has not however, in the view of the Court, adduced clear evidence which establishes a link between the nonylphenols found in the waters of the river and the Orion (Botnia) mill. [. . . . ] The Court therefore concludes that the evidence in the record does not substantiate the claims made by Argentina on this matter.”56

 Dioxins and furans claim: “The Court considers that there is no clear evidence to link the increase in the presence of dioxins and furans in the river to the operation of the Orion (Botnia) mill.”57

 Biodiversity claim: “The Court has not, however, found sufficient evidence to conclude that Uruguay breached its obligation to preserve the aquatic environment including the protection of its fauna and flora.”58 And “[t]he record rather shows that a clear relationship has not been established between the discharges from the Orion (Botnia) mill and the malformations of rotifers, or the dioxin found in the sábalo fish or the loss of fat by clams reported in the findings of the Argentine River Uruguay Environmental Surveillance (URES) programme.”59

While it might be easy to miss, it is important to note that the Court does not make factual findings with regard to claims other than the phosphorous claim.60 The Court comes close at times to making a factual conclusion.61 For instance, it points to certain reports that are inconsistent with Argentina’s submissions.62 But such excursions by the Court are not an affirmative determination of fact. They do not definitively conclude that dissolved oxygen levels in the River Uruguay on the critical date were at or below a certain level.63 Given the sophistication of the jurists on the Court, it is hard to avoid that this (lack of a) conclusion is purposeful. The conclusions in the Pulp Mills case therefore turn entirely on the question of burdens of proof. The Court assigned the burden

55. Id. at 97-98. 56. Id. at 98-99. 57. Pulp Mills, 2010 I.C.J. at 99. 58. Id. at 100. 59. Id. 60. Id. at 96. 61. Id. at 93. 62. Id. 63. Pulp Mills, 2010 I.C.J. at 93. 114 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 to Argentina.64 Argentina did not meet its burden.65 It therefore did not make out a claim for relief.66 Put another way, had the burden of proof been on Uruguay in this case, Uruguay likely would have lost the case. The Court precisely refused to make any affirmative factual findings, as we have now seen. Consequently, the burden of proof, and the burden of proof alone, was dispositive of the factual questions in the case. The importance of burdens of proof in the Court’s jurisprudence is not unique to the Pulp Mills case. Rather, it is a generally accepted proposition that the assignment of the burden of proof is a central feature of the factfinding process.67 Pulp Mills is a vivid illustration of how this principle is applied even in cases in which the Court appears to suggest that it will make findings of fact on its own recognizance rather than decide the case on the relative strength of the evidence submitted by the parties.68 Even in this context, burden is inescapable. How then does the Court determine who has the burden of proof? The rules on burdens of proof are generally well- established and straight forward. The Court in the Pulp Mills case explained as follows: “in accordance with the well-established principle of onus probandi incumbit actori, it is the duty of the party which asserts certain facts to establish the existence of such facts.”69 It went on to explain that “the Applicant should, in the first

64. Id. at 71. 65. Id. at 93, 97-8, 98-99, 99, 100. 66. Id. 67. Genocide Case (Croat. v. Serb.), Judgment, 2015 I.C.J. 3, 73-75 (Feb. 3) (discussing the importance of burden of proof); Maritime Delimitation in the Black Sea (Rom. v. Ukr.), Judgment, 2009 I.C.J. 61, 86 (Feb. 3) (same); Sovereignty over Pedra Branca/Pulau Batu Puteh, Middle Rocks and South Ledge (Malay. v. Sing.), Judgment, 2008 I.C.J. 12, 31 (May 23); Genocide Case (Bos. & Herz. v. Serb.), Judgment, 2007 I.C.J. 43, 127-29 (Feb. 26) (same); Military and Paramilitary Activities in and against Nicaragua (Nicar. v. U.S.), 1984 I.C.J. 392, 437 (Nov. 26). 68. Pulp Mills, 2010 I.C.J. at 72-73. 69. Id. at 71; Temple of Preah Vihear (Cambodia v. Thai.), Judgment, 1962 I.C.J. 6, 15- 16 (June 15); see CHENG, supra note 12, at 327 (establishing burden of proof as a general principle of law); ROBERT KOLB, THE INTERNATIONAL COURT OF JUSTICE 931 (Hart Publishing 2013) (discussing assignment of burden of proof in International Court of Justice jurisprudence); Anna Riddell, Evidence, Fact-Finding, and Experts, in THE OXFORD HANDBOOK OF ADJUDICATION 858, 858-59 (Cesare Romano et al. eds., 2014) (discussing the principle in current practice); DURWARD V. SANDIFER, EVIDENCE BEFORE INTERNATIONAL TRIBUNALS 127 (University Press of Virginia 1939) (discussing the same principle in historical practice). 2020] TRUTHS IN TRANSLATION 115 instance, submit the relevant evidence to substantiate its claims.”70 It went on that “[t]his does not, however, mean that the Respondent should not co-operate in the provision of such evidence as may be in its possession that could assist the Court in resolving the dispute submitted to it.”71

2. Presumptions Shift the Basic Rule The Court in some circumstances is willing to soften this operation of the burden of proof as a substitute for fact finding. The Diallo case is one example of such a softening.72 In Diallo, Guinea brought a case against the Democratic Republic of Congo for the mistreatment of one of its nationals at the hands of the Congolese authorities.73 Guinea alleged as one part of its case that the Democratic Republic of Congo had failed to provide Diallo with basic procedural guarantees.74 Given the alleged Kafkaesque nature of these infractions, it would have been close to impossible for Mr. Diallo (and thus Guinea) to provide documentary proof of these infractions. The Court in this context appeared willing to adjust the ordinary operation of burdens of proof against the moving party.75 It explained that the burden of proof “varies according to the type of facts which it is necessary to establish for the purposes of the decision of the case.”76 Specifically, “where, as in these proceedings, it is alleged that a person has not been afforded, by a public authority, certain procedural guarantees to which he was entitled, it cannot as a general rule be demanded of the Applicant that it prove the negative fact which it is asserting.”77 The Court did not however reverse the burden of proof.78 Cryptically, it concluded that “[i]t is for the Court to evaluate all the evidence produced by the two Parties and duly subjected to

70. Pulp Mills, 2010 I.C.J. at 71. 71. Id. 72. Ahmadou Sadio Diallo (Guinea v. Dem. Rep. Congo), Judgment, 2010 I.C.J. 639 (Nov. 30) [hereinafter Diallo]. 73. Id. at 645. 74. Id. at 659-60. 75. Compare Pulp Mills, 2010 I.C.J. at 71 (setting out the ordinary operation of burdens as quoted above), with Diallo, 2010 I.C.J. at 660 (softening the burden of proof). 76. Diallo, 2010 I.C.J. at 660. 77. Id. at 660-61. 78. Id. at 661. 116 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 adversarial scrutiny, with a view to forming its conclusions.”79 As “when it comes to establishing facts such as those which are at issue in the present case, neither party is alone in bearing the burden of proof.”80 Here then, the Court appears to say that it would make the kinds of affirmative findings of fact that it avoided in Pulp Mills.81 It would no longer rule that it “has not, however, found sufficient evidence to conclude that” the Respondent breached its obligations.82 Applied more broadly, it would have to hold that on the whole, the record requires the making of a finding of fact in situations like Diallo. Once such a fact of fact has been made, it would then in turn be possible to rule on the question whether a respondent has (or has not) breached an international legal obligation. It is tempting to view the Diallo decision as affecting standards of proof rather than burdens of proof. As discussed in the next section, the standard of proof governs the amount and quality of evidence needed to support a finding. The Diallo Court appears to be saying that it will permit Guinea to establish a fact with less evidence, or evidence less directly in support of its contentions, than it otherwise would have.83 This conclusion would not affect burdens of proof. The Court, however, does more than that.84 It adjusted the burden of proof by introducing a presumption into the operation of burdens.85 The presumption it introduced is that governmental agencies keep records of their own, ordinary application of procedural guarantees in due course.86 This presumption does not just substitute for a piece of evidence. Rather, it interacts directly with the process of fact finding itself: the presumption affects how the Court engages in its fact-finding mission by requiring it to make a factual finding unlike the posture in Pulp Mills even in the face of no other record evidence supporting Guinea’s contention rather

79. Id. 80. Id. 81. See generally Pulp Mills (Arg. v. Urug.), Judgment, 2010 I.C.J. 14 (Apr. 20). 82. Id. at 100. 83. Diallo, 2010 I.C.J. at 661. 84. Id. 85. See id. (“A public authority is generally able to demonstrate that it has followed the appropriate procedures”). 86. Id. 2020] TRUTHS IN TRANSLATION 117 than what facts it finds.87 In Pulp Mills a presumption would have been one tool to satisfy a burden of proof but the Court could still have refused to make a finding at all by reference to applicable burdens.88 Diallo invokes a presumption to deprive the Court of this option to return nonsuit by ordinary operation of burdens of proof.89 Diallo reveals the fundamental importance of presumptions: presumptions can alter the process of fact finding.90 Presumptions are applied not to the record but to the manner of application of burdens and standards of proof. They can alter even the most basic features we take for granted—the rule that the moving party must shoulder the burden of proving the facts upon which it relies or fail in making out its factual case. This central importance of presumptions means that they deserve significant scrutiny as they are at the heart of the fact-finding process of international justice— they define how international justice knows or understands the world in which it operates.

3. Burdens, Presumptions, and (Post-)Truth The discussion so far should readily bring to light what our problems are if we seek to rely on the international law of evidence as the rule of law toolkit to address the post-truth syndrome. First, it might certainly seem intuitive to say that the person who wants to make a factual submission must prove the point to the satisfaction of its audience.91 However, if we apply the Pulp Mills matrix to the question of climate change (either existential threat or “Chinese hoax”), one might think twice.92 We hear people

87. Pulp Mills, 2010 I.C.J. at 100; FREDERIC G. SOURGENS, KABIR DUGGAL & IAN LAIRD, EVIDENCE IN INTERNATIONAL INVESTMENT ARBITRATION 291 (2018) (Evidentiary Principles §1(1)). 88. Pulp Mills, 2010 I.C.J. at 100. 89. Diallo, 2010 I.C.J. at 661. 90. Id. 91. Pulp Mills, 2010 I.C.J. at 71. 92. Louis Jacobsen, Yes, Donald Trump Did Call Climate Change a Chinese Hoax, POLITIFACT (June 3, 2016), https://www.politifact.com/truth-o- meter/statements/2016/jun/03/hillary-clinton/yes-donald-trump-did-call-climate- change-chinese-h/ [https://perma.cc/7SBX-H77J]; Andrew O’Reilly, Trump Hedges on Climate Change Denial in Wake of Hurricane Michael; Says 'There Is Something There,’ FOXNEWS.COM (Oct. 15, 2018), https://www.foxnews.com/politics/trump-hedges-on- climate-change-denial-in-wake-of-hurricane-michael-says-there-is-something-there [https://perma.cc/Q4GB-QRMH]. 118 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 engaged in the post-truth discourse on climate change say “I am not a climate scientist, but . . . ” and then raise cold winter temperatures as a reason to doubt “global warming.”93 This is not unlike the Pulp Mills decision: in it, the Court seems to say “I am not an environmental scientist, but . . . . ” Moreover, to place the burden on the party seeking to assert a human link to environmental degradation in a mechanical fashion, as Pulp Mills appears to have done, is itself a tool all too familiar to any person watching Fox News.94 Second, the embedding of presumptions in the burden of proof we encountered in Diallo, too, might look reasonable. But again, the question arises: whose presumption governs? Presumptions draw on narrative, and if narrative affects the manner in which we apply burdens of proof, we are again perilously close to falling into the trap of post-truth discourse rather than to providing a tool to remedying it. The rule of law, in other words, must do better than this use of burdens of proof to live up to its aspirations to be a post-truth antidote. A step to addressing this problem is the standard of proof.

B. Standard of Proof Global justice in general requires the moving party to convince the “inner conviction” of the judge of the facts asserted.95 Confusingly, this global standard has been compared to the preponderance standard and the clear and convincing evidence

93. Rebecca Onion, How to Stop Falling for the “I’m not a Scientist” Trap, SLATE (Nov. 29, 2018), https://slate.com/technology/2018/11/donald-trump-not-a-scientist- climate-denialism-rhetoric.html [https://perma.cc/3P2K-7CPE]; Abby Smith, Trump Says World Will ‘Start Getting Cooler’ as Biden Criticizes Him as a ‘Climate Arsonist’, WASH. EXAMINER (Sept. 14, 2020), https://www.washingtonexaminer.com/policy/energy/trump-says-world-will-start- getting-cooler-as-biden-criticizes-him-as-a-climate-arsonist [https://perma.cc/9T7L- ULZY]. 94. Pulp Mills, 2010 I.C.J. at 71; Michael Guillen, Physicist: Don’t Fall for the Argument about ‘Settled Science’, FOX NEWS (Jan. 21, 2019), https://www.foxnews.com/opinion/physicist-dont-fall-for-the-argument-about-settled- science [https://perma.cc/XM6L-GNB7]. 95. Armed Activities on the Territory of the Congo (Dem. Rep. Congo v. Uganda), Dissent, 2005 I.C.J. 361, 361 (Dec. 19) (dissenting opinion by Kateka J.). 2020] TRUTHS IN TRANSLATION 119 standard.96 Pragmatically, the standard appears to oscillate between the two in requiring more than a mere preponderance but requiring less than clear and convincing evidence without precise demarcation as to the amount of proof required.97 Given this standard, it is intriguing to see how the Court addressed the question whether all cases are created equal in terms of the applicable standard of proof when it was asked to do so head-on. The issue was litigated in the 1949 .98 Corfu Channel involved a claim by the against Albania with regard to deaths and significant damage done to vessels by mines laid in the North Corfu Straight on October 22, 1946.99 The United Kingdom asserted that it had recently cleared mines in the straight in 1944 and re-swept the straight in 1945.100 On its passage in October 1946, the British destroyer Saumarez nevertheless hit a mine on its sailing through the straight.101 The Volage attempting to rescue the Saumarez also struck a mine.102 Both vessels were severely damaged as a result.103 The incident caused the death of 44 sailors and injured an additional 42.104 The United Kingdom argued that Albania had colluded with the Yugoslav navy to lay the mines in question to impede future (British) traffic through the channel in violation of international law.105 The United Kingdom thus alleged as one of its theories of the case that the mines had been laid by Yugoslav vessels with the express permission of the Albanian government.106 As proof, the United Kingdom submitted testimony from LCdr. Kovacic, formerly of the Yugoslav Navy. 107 He testified he witnessed two vessels loading mines at Sibenik (Yugoslavia) and leaving port, and

96. See Kevin M. Clermont, Standards of Proof in Japan and the , 37 CORNELL INT’L L.J. 263, 266 (2004) (situating the standard in the context of US standards of proof). 97. Kolb, supra note 69, at 944. 98. Corfu Channel (U.K. v. Alb.), Judgment, 1949 I.C.J. 4 (Apr. 9). 99. Id. at 10-11. 100. Id. at 10. 101. Id. at 12. 102. Id. at 13. 103. Corfu Channel, 1949 I.C.J. at 10. 104. Id. at 10. 105. Id. at 16. 106. See id. at 15. 107. Id. at 8. 120 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 then observed their return “a few days after the occurrence of the explosions.”108 The case posed delicate questions regarding the applicable standard of proof. Yugoslavia was not a party to the Corfu Channel proceedings.109 It therefore did not present evidence (or examined LCdr. Kovacic).110 This meant that there was reasonably little in evidence to counter LCdr. Kovacic’s account of events as the party with the evidence in question was (legitimately) absent. An ordinary application of a standard of proof would thus suggest that some credible evidence carries the day over no evidence—i.e., that the United Kingdom would prevail on its claim. Given the procedural posture of the case, this was deeply problematic not the least of which because it would imply liability for a third and absent party, Yugoslavia.111 The Court dealt with the question by going into further detail regarding the applicable standard of proof. It ruled: “[w]ithout deciding as to the personal sincerity of the witness Kovacic, or the truth of what he said, the Court finds that the facts stated by the witness from his personal knowledge are not sufficient to prove what the United Kingdom Government considered them to prove.”112 The Court explained that “[a] charge of such exceptional gravity against a State would require a degree of certainty that has not been reached here.”113 Narrowly, a grave allegation of wrongdoing must be supported by credible evidence that requires only limited inferences: seeing Yugoslav vessels loaded with mines leave port at Sibenik and returning after a voyage is consistent with the laying of mines in the channel. But it does not directly suggest the laying of mines in the channel—the vessels in question plausibly could have had an entirely different mission. At the very least, the witness would have had to place the vessels in the channel acting in a manner consistent with laying mines. The Court’s holding thus altered the standard of proof by requiring not more proof (i.e., 15 additional eye witnesses) but a

108. Corfu Channel, 1949 I.C.J. at 16. 109. Id. at 17. 110. Id. at 17. 111. For a discussion of why this would be problematic from a procedural point of view, see Monetary Gold Removed from Rome in 1943 (It. v. Fr., U.K. & U.S.), Judgment, 1954 I.C.J. 19, 18-19 (June 15). 112. Corfu Channel, 1949 I.C.J. at 16. 113. Id. at 17. 2020] TRUTHS IN TRANSLATION 121 different kind of proof that permits the inference in question with more immediacy. The Court continues to apply this heightened evidentiary requirement to this day.114 The key to understanding this application of the standard of proof in Corfu Channel (and later ICJ jurisprudence) is to contrast it with the manner in which the Court ultimately ruled against Albania in the case. The Court ruled that Albania did not itself lay the mines.115 Further, the Court did not find in favor of the United Kingdom with regard to the connivance theory.116 Rather, the Court ruled that the presence of the mines in Albanian waters permitted a relaxation of standards of proof as Albania was best positioned to provide an explanation for their presence.117 The Court further inferred knowledge of the presence of the mines from the lack of protest against mining in its waters after the minefield in question had officially been recorded and the ease with which geography permitted Albania to observe the straight from its coastline.118 The Court expressly placed this discussion of the standard of proof in the context of its discussion of burdens of proof.119 Thus, the Court ruled that the case did not shift the burden of proof to Albania but affected the standard of proof.120 This standard of proof was placed directly in the context of more liberal recourse to inferences and circumstantial evidences given the facts at bar and the relative access to evidence enjoyed by the parties.121 Viewed together, the Corfu Channel decision thus heightened and lowered the standard of proof in the same decision with regard to the same question—who was responsible for the laying of mines in the straight? This would on its face appear to be further proof of a lack of evidentiary rigor by the Court. The same type of allegation should require any factfinder to apply the same standard of proof,

114. See Application of Convention on Prevention and Punishment of Crime of Genocide (Croat. v. Serb.), Judgment, 2015 I.C.J. 3, ¶¶ 178-79 (Feb. 3); see also Application of Convention on Prevention and Punishment of Crime of Genocide (Bosn. & Herz. v. Serb. & Montenegro), Judgment, 2007 I.C.J. 43, ¶ 209 (Feb. 26). 115. Corfu Channel, 1949 I.C.J. at 16. 116. See id. at 16-18. 117. Id. at 18. 118. Id. at 18-20. 119. See id. at 18. 120. Corfu Channel, 1949 I.C.J. at 18. 121. Id. 122 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 if not between different disputes, at least in the same judgment. The oscillation on point is therefore disquieting. Despite this facial incongruence, it is nevertheless possible to tease out a common denominator for the Court’s fact-finding process. The application of the standard of proof followed a common factor: presumptions. Specifically, the process becomes uniform when we ask how could the Court draw inferences consistent with its expectation of how states (ought to) behave? On the one hand, one would not expect a state to connive in breaking the law with a particular partner.122 Such an inference therefore requires more direct proof.123 On the other hand, one would expect a state to know of, and take affirmative action with regard to, notorious minelaying by a third party in a strategic part of .124 The standard of proof thus described the manner in which the Court was willing to draw facially opposite inferences on the basis of a common set of presumptions about plausible state action in the case.125 In classical international adjudication, standards of proof, therefore, only make sense when they are evaluated in the broader context of inferences and presumptions in the context of which they operate. This use of standards of proof is certainly helpful in addressing the concerns raised in the context of post-truth discourse in the previous section. We see that burdens are not nakedly imposed. Rather, they are inherently sensitive to context. In this context, burdens and standards operate so as to take the sting out of some of the invective underlying the post-truth discourse: the fear of the hoax or connivance.126 To place the discussion back in the context of climate change deniers on television, the use of standards of proof in Corfu Channel would be highly skeptical of anyone ascribing motive to scientists warning of potential dangers.127 This may thus help to bring parties back to the table to have an honest discussion rather than hurling narrative-based innuendo.

122. See id. at 16-18. 123. See id. at 16. 124. See id. at 18-20. 125. See Corfu Channel, 1949 I.C.J. at 20. 126. See id. at 16-18; see also Jacobson, supra note 92. 127. See Corfu Channel, 1949 I.C.J. at 16-18. 2020] TRUTHS IN TRANSLATION 123

More centrally, the use of standards of proof as evidenced by Corfu Channel is an antidote to strategies glorified in the “Rules of Radicals” manuals (or their current understandings, at any rate).128 These manuals suggest as one of the first steps of radical discourse to use such innuendo to discredit one’s opponent and create an atmosphere of fear or anxiety.129 Sadly, it appears that in a post- truth society, the use of such techniques has become common.130 Or, as one leading international legal scholar recently put it, the “manipulation of anxiety is a standard technique of social control.”131 The use of standards of proof would minimize manipulations of anxiety by requiring affirmative proof of such claims rather than mere innuendo thus taking at least some of the wind out of the sails of radical discourse participants. However, as much as standards of proof appear helpful in resolving some of the problems we encounter in the post-truth discourse, it again highlights the importance of narratives. The standard of proof is highly narrative dependent because it, too, relies upon presumptions. The question then becomes—how do we justify the use of presumptions? Could they not be easily replaced by other, more corrosive presumptions premised in alternative narratives? The point is important as it is a central tenet of Rules for Radicals to “[n]ever go outside of the experience of your people.”132 The link between experience, narrative, and presumption is thus as central to poisoning or radicalizing discourse in post-truth society as it is a potential antidote to it. In other words, the law governing presumptions is going to be instrumental in understanding the rule of law answer to narrative- based discourse—or in revealing that rule of law has no such answer.

128. SAUL D. ALINSKY, RULES FOR RADICALS (1971); see generally Noam Cohen, Know Thine Enemy, N.Y. TIMES (Aug. 22, 2009), https://www.nytimes.com/2009/08/23/weekinreview/23alinsky.html [https://perma.cc/E2HB-WZ6J]. 129. ALINSKY, supra note 128, at 99. 130. See Michael Tackett & Maggie Haberman, Trump Once Said Power Was About Instilling Fear. In That Case, He Should Be Worried, N.Y. TIMES (Feb. 4, 2019), https://www.nytimes.com/2019/02/04/us/politics/fear-trump.html [https://perma.cc/9JLG-CJEQ] (discussing the use of fear). 131. W. MICHAEL REISMAN, THE QUEST FOR WORLD ORDER AND HUMAN DIGNITY IN THE TWENTY-FIRST CENTURY 67 (2012). 132. ALINSKY, supra note 128, at 127. 124 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1

C. Burdens and Standards Cannot Operate Without Presumptions The analysis of the leading forms of international justice has established that burdens and standards of proof are the foundational building blocks for fact finding in international dispute resolution. The discussion so far has also shown that these burdens and standards of proof do not operate mechanically or blindly. Rather, they operate in close interaction with the law of inferences and in particular the law of evidentiary presumptions. These presumptions determine how justice knows the facts and appreciates how the burden of proof is to be applied in any given case to resolve hard cases. This means that evidentiary presumptions and their relationship to inferences constitute the core operational code of international dispute resolution—the principle that determines how to resolve factual disputes following the international rule of law.

III. INFERENCES AND PRESUMPTIONS The role of inferences and presumptions becomes more readily apparent still when one dissects how proof in international dispute resolution actually progresses. This section begins with the much-touted preferences for direct evidence and how such evidence interacts with proof by inferences. Specifically, it explains how proof by inference in most (if not all) cases is actually dispositive of the factual questions to be resolved by a court or tribunal. The section then looks at the relationship between inferences and presumptions. It submits that presumptions are a certain kind of general inference. It then outlines how these general presumptions overlap with specific inferences in order to dispose of contested factual questions in international adjudication and international arbitration.

A. Proof by Direct Evidence It is almost a knee-jerk reaction to prefer direct evidence over circumstantial evidence. This knee-jerk reaction has been hardwired into our TV viewing habits: we have been fed “your case is entirely circumstantial” as good lawyerly argument to discredit 2020] TRUTHS IN TRANSLATION 125 an assertion from Legal Eagles to Law & Order to Bull.133 This knee- jerk reaction is to a certain extent present in international dispute resolution, but it must be placed in context of the central work inferences play when assessing direct evidence. International justice—like our average television viewer— has a preference for direct evidence. As one recent work restating the rules of evidence in the international investment arbitration context has noted, arbitrators “when possible shall make findings of fact by means of direct evidence.”134 This preference for direct evidence also underlies international adjudication at the International Court of Justice.135 While we are used to the distinction between direct evidence and circumstantial evidence from a steady diet of courtroom dramas, what constitutes direct evidence is far less clear. According to a classic definition, direct evidence “proves a fact without an inference or presumption and which in itself, if true, establishes that fact.”136 When asked, we might think of an eyewitness.137 Or we might, mistakenly it turns out, think of real evidence like a gun or the smudge of blood left behind on a broken ground floor window. (This real evidence requires inferences to establish a fact—the gun was used in the crime, the defendant held it, the blood on the window was left by the assailant at a certain time etc.).138 As the discussion below will show, witness evidence is far more controversial in the international setting than it is in the US setting. International justice instead relies predominantly upon a different kind of evidence as its primary source of direct evidence:

133. See Jim Cash & Jack Epps, Jr., script authors, Legal Eagles (1986) Quotes, IMDB, https://www.imdb.com/title/tt0091396/quotes [https://perma.cc/8RLX-83GE] (last visited Oct. 4, 2020); see also Robert Nathan, script author, Law & Order, The Secret Sharers (1991) Quotes, IMDB, https://www.imdb.com/title/tt0629460/quotes [https://perma.cc/Q5LC-TWB4] (last visited Oct. 4, 2020). 134. SOURGENS ET AL., supra note 87, at 292. 135. Certain Activities Carried out by Nicaragua in the Border Area (Costa Rica v. Nicar.), Judgment, 2015 I.C.J. 665, 733 (Dec. 16) (noting the absence of direct evidence submitted by Nicaragua as a significant factor in its decision). 136. Kevin J. Heller, The Cognitive Psychology of Circumstantial Evidence, 105 MICH. L. REV. 241, 248 (2006) (quoting BARBARA E. BERGMAN & NANCY HOLLANDER, WHARTON’S CRIMINAL EVIDENCE §1“8” (1997)). 137. Heller, supra note 136, at 248. 138. Id. at 251. 126 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 contemporaneous documentary evidence.139 (This evidence is not formally primary in the US setting as it is almost by definition hearsay evidence, i.e., an out of court of statement submitted for the truth of the matter asserted.)140 This section will outline the key differences between the treatment of this preferred form of evidence and witness evidence.

1. Contemporaneous Documentary Evidence Contemporaneous documentary evidence is the gold standard for global dispute resolution. This standard has been embraced in international adjudication,141 and plays a central role in international arbitration.142 It has come to stand for the most immediate and most trustworthy kind of evidence upon which a decision may be based. This trustworthiness is partly due to the types of questions international justice is asked to resolve. Take, for example, two staples of the Court’s Jurisprudence: border delimitations and maritime claims.143 When the International Court of Justice is asked to delimit a border, or determine a maritime claim, one of the issues will be one party’s historical claim to the particular area.144 This historical claim would be difficult to establish by means of witness testimony as the witness would have to have knowledge of prevailing circumstances centuries ago.145 Witness evidence in this context is therefore far less valuable as a practical matter. Therefore, directly establishing such historical claims will be contingent upon historical documents. These documents will most

139. SOURGENS ET AL., supra note 87, at 299. 140. FED. R. EVID. 801(c). Of course, contemporaneous documents authored by a party opponent are defined as “not hearsay” for purposes of the federal rules. FED. R. EVID. 801(2). 141. Simone Halink, All Things Considered: How the International Court of Justice Delegated its Fact Assessment to the in the Armed Activities Case, 40 N.Y.U. J. INT’L L. & POL. 13, 22 (2008). 142. SOURGENS ET AL., supra note 87, at 299. 143. See John R. Crook, The International Court of Justice and Human Rights, 1 NW. J. INT’L HUM. RTS. 1, 3 (2004) (“More than half its cases have involved disputes over land frontiers and maritime boundaries”). 144. Brian Taylor Sumner, Territorial Disputes at the International Court of Justice, 53 DUKE L.J. 1779, 1789 (2004). 145. See id. at 1789-90 (discussing the importance of history in border delimitations). 2020] TRUTHS IN TRANSLATION 127 clearly establish the state of affairs at the time of the making of the document. The most frequent such document is a map.146 Maps play an outsized role in international dispute resolution as a direct evidence of the respective (historical) claims of the parties. Even with such maps, one needs to be careful. The Court in the Frontier Dispute case explained that maps “of themselves, and by virtue solely of their existence, cannot constitute a territorial title, that is, a document endowed by international law with intrinsic legal force for the purpose of establishing territorial rights.”147 In other words, maps are evidence of a fact, not an instrument of title.148 They are certainly probative as direct evidence of contemporaneous views of their drafters. Their probative force will vary, however, depending upon the accuracy and technical reliability with which they have been made.149 In this sense, maps are not different from an eyewitness who after all is only as good as her eyesight permits her to be. Further their credibility may be called into question even if they have significant probative force. The question why the map was drawn up looms large in this context.150 Further, it might even be proved that the map is not authentic to a relevant period.151 Finally, even if one establishes a credible historical claim on the basis of a map, this is not the only factor to be considered as a matter of law.152 Other such factors considered by the Court include geography (for which different, contemporary maps may be more instructive than historical ones), cultural identity, and, centrally, earlier negotiated settlements of boundary questions.153

146. See, e.g., Temple of Preah Vihear (Cambodia v. Thai.), 1962 I.C.J. 6, 9-32 (June 15) (discussing maps); Maritime Dispute (Peru v. Chile), 2014 I.C.J. 3, 64 (Jan. 27) (same); Maritime Delimitation in the Black Sea (Rom. v. Ukr.) , 2009 I.C.J. 61, 83 (Feb. 3) (same). 147. Frontier Dispute (Burk. Faso v. Mali), Judgment, 1986 I.C.J. 554, 582 (Dec. 22); Territorial and Maritime Dispute (Nicar. V. Colom.), Judgment, 2012 I.C.J. 624, 661-2 (Nov. 19) (quoting Burk. Faso v. Mali, 1986 I.C.J. at 582). 148. Burk. Faso v. Mali, 1986 I.C.J. at 582. 149. Id.; see also Frontier Dispute (Burk. Faso v. Niger), 2013 I.C.J. 44, 76 (Apr. 16) (relying on the 1986 Frontier Dispute case for the same point); Peru v. Chile, 2014 I.C.J. at 64 (maps focus on the locations inapposite to the litigation). 150. Burk. Faso v. Mali, 1986 I.C.J. at 583. 151. See id. at 584 (noting the curious disappearances of maps appended to treaties and the presence of an apparent wealth of cartographic evidence “for a region which is nevertheless described as being partly unknown”). 152. See Sumner, supra note 144, at 1779-80; see, e.g., Rom. v. Ukr., 2009 I.C.J. at 83. 153. See Sumner, supra note 144, at 1780; Rom. v. Ukr., 2009 I.C.J. at 83. 128 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1

Many international adjudications turn critically on the diplomatic engagement between two states. Fact finding in these cases relies on two types of contemporaneous documents: (1) diplomatic notes154 and (2) contemporaneous governmental documents relating to the conduct of foreign affairs.155 The East Greenland case is an early example of the use of direct evidence in such cases.156 The East Greenland case in relevant part concerned the question whether Norway had ceded its claim to sovereignty over Greenland to Denmark.157 As evidence of such a cession, Denmark submitted that Norway’s Foreign Minister, Mr. Ihlen, had made a promise to cede Norway’s claim to Greenland in exchange for Danish concessions with regard to Norwegian claims to the island of Spitzbergen.158 The Danish case rested on two distinct sets of documents as direct evidence for the promise.159 It relied in the first place upon Ihlen’s own protocol entries regarding the conversations with his Danish counterpart.160 It then placed these protocols into context of diplomatic exchanges between Denmark and Norway at the time by submitting its own documents preparing the conversation with the Norwegian foreign minister.161 It finally looked to later diplomatic correspondence between Denmark and Norway, in which Denmark repeated claims to Greenland and Norway refused to ratify the earlier remarks.162 The Court relied as direct evidence upon these contemporaneous documents.163 It considered first and foremost the protocol entries of Mr. Ihlen made at or shortly after the

154. See, e.g., Legal Status of Eastern Greenland (Den. v. Nor.), 1933 P.C.I.J. (ser. A/B) No. 53 (Apr. 5); Sovereignty over Palau Litigan and Palau Sipadan (Indon. v. Malay.), 2001 I.C.J. 575, 592 (Oct. 23); Territorial and Maritime Dispute (Nicar. v. Hond.), 2007 I.C.J. 659, 737 (Oct. 8); Sovereignty Over Pedra Branca (Malay. v. Sing.), 2008 I.C.J. 12, 27 (May 23); Maritime Delimitation in the Indian Ocean (Som. v. Kenya),2017 I.C.J. 3, 15 (Feb. 2). 155. Alleged Violations of Sovereign Rights and Maritime Spaces in the Caribbean Sea (Nicar. v. Colo.), Judgment, 2016 I.C.J. 3, 28 (Mar. 17). 156. Den. v. Nor., 1933 P.C.I.J. at 37, 70. 157. Id. at 71. 158. Id. 159. Id. at 37, 70. 160. Id. at 70. 161. Den. v. Nor., 1933 P.C.I.J. at 70. 162. Id. at 37. 163. Id. at 70-72. 2020] TRUTHS IN TRANSLATION 129 discussions with the Danish foreign minister.164 To generalize, the Court relied upon contemporaneous governmental documents memorializing foreign affairs exchanges as primary means of direct proof. It then also considered written diplomatic exchanges as evidence to confirm the earlier oral exchanges. The same pattern of a crisscrossing reliance on government documents and diplomatic communications still remains in place in contemporary dispute resolution before the Court.165 For instance, Alleged Violations of Sovereign Rights and Maritime Spaces in the Caribbean Sea involved a critical question when Nicaragua notified Colombia of an alleged violation of its sovereign rights.166 Colombia submitted a diplomatic note sent by Nicaragua after the institution of the proceedings before the Court, together with the absence of previous notes, as direct evidence of this date.167 The Court looked to Colombian governmental documents to establish that Nicaragua had made oral demands of Colombia at an earlier time.168 On this basis, the Court concluded that “Given the public statements made by the highest representatives of the Parties,” including presidential offers of negotiations extended by both presidents, “Colombia could not have misunderstood the position of Nicaragua over such differences.”169 Here, again, the use of contemporaneous governmental documents about foreign affairs proved dispositive direct evidence. The Court in this case used these documents to explain, and provide context for, later diplomatic exchanges.170 The case thus confirms the use of government documents and diplomatic correspondence as a preferred means of proof through direct evidence. Despite this preference, a word of caution is needed: ultimately, even contemporaneous documentary evidence is viewed through the lens of presumptions. Thus, the Court typically does not have a particular desire to get “behind” contemporaneous foreign affairs related documents by looking at the documents of

164. Id. at 70-72. 165. Alleged Violations of Sovereign Rights and Maritime Spaces in the Caribbean Sea (Nicar. v. Colo.), Judgment, 2016 I.C.J. 3, 26 (Mar. 17). 166. Id. 167. Id. at 28. 168. Id. at 32-33. 169. Id. at 31, 32-33. 170. Nicar. v. Colo., 2016 I.C.J. at 31, 32-33. 130 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 other governmental departments. The issue most embarrassingly came up in the Corfu Channel case already discussed above.171 One of the questions the Court needed to establish was whether British passage through the straight would qualify as “innocent passage” (broadly, without belligerent intent) as the UK government asserted.172 If the Court established that British passage was not innocent, then the U.K. would lose its claim for the violation of its rights to innocent passage by the mining efforts in the straights.173 The problem for the Court was that the Royal Navy had a document—known as Document XCU—that outlined the reasons for British passage through the straights, namely, reconnaissance of Albanian naval defenses.174 Albania had requested production of the document.175 The Court had ordered it be produced.176 After much back and forth, the UK refused to produce it.177 And yet—the Court did not draw the inference from the non-production of Document XCU that British passage was not innocent.178 Why? The question cannot definitively be answered. The Court on its face appeared to reason that the documents could not have contained any damning information as the British vessels did not fire upon Albanian positions despite having struck mines.179 This conclusion seems logically contorted as it does not follow that a reconnaissance mission under the circumstances would require the firing of shots if the British vessels in question struck mines (as opposed to being fired upon from land).180 The British navy, in other words received the benefit of the doubt in circumstances that on their face would have justified significant suspicion. What then really motivated the Court? It would appear that the Court is loath to second guess the “secret” motivations of states even on the basis of contemporaneous documentary evidence. In

171. Corfu Channel (U.K. v. Alb.), Judgment, 1949 I.C.J. 4, 4 (Apr. 9). 172. Id. at 10. 173. See id. at 28. 174. Jeffrey Dunoff & Mark Pollack, International Judicial Performances and the Performance of the International Court, in THE PERFORMANCE OF INTERNATIONAL COURTS AND TRIBUNALS 261, 279 (Theresa Squatrito et al. eds., 2018). 175. Corfu Channel, 1949 I.C.J. at 32. 176. Id. 177. Id.; Dunoff & Pollack, supra note 174 (describing discussions relating to the non- production of documents in Corfu Channel); REISMAN, supra note 131, at 45 (same). 178. Corfu Channel, 1949 I.C.J. at 32. 179. Id. 180. See id. 2020] TRUTHS IN TRANSLATION 131 more political terms, the Court may very well be unwilling to rub the noses of the world community in deeply problematic areas of foreign policy. To second guess states’ motives is to assign more than a modicum of blame.181 In Corfu Channel, the Court would have assigned such blame to a Western power (the U.K.) with regard to potentially belligerent intent towards an country (Albania).182 Such an exercise is a dangerous proposition for a Court that in the end relies upon the consent of disputing parties across the geopolitical divides to its jurisdiction.183 Further, it is potentially unhelpful to the pacific resolution of international disputes. Guilt and blame beget more deflections of blame without necessarily permitting the resolution of factual disputes on common ground.184 In Corfu Channel, such blame would have brought forth claims that Albania (and the Soviet bloc) was the true aggressor and that Britain was merely “defending itself” and its rights of navigation.185 The post-truth era has made this phenomenon the more readily apparent: facts can be brushed aside with alternative facts when arguments of blame and identity collide.186 To shy away from finding facts that would elicit an identity-based (rather than a strictly factual) response might therefore be a condition sine qua non for pacific dispute settlement to be possible more generally. Pacific dispute settlement mechanisms can do so by invoking presumptions. And this use of presumptions allows pacific dispute settlement to avoid the taboo of sliding from legal dispute into political confrontation.187

181. Jean Bethke Elshtain, The Perils of Legal Moralism, 20 J.L. & POL. 549, 564 (2004) (discussing the problem through the lens of legalistic moralism). 182. LCDR Weston D. Burnett, Mediterranean Mare Clausum in the Year 2000?: An International Law Analysis of Peacetime Military Navigation in the Mediterranean, 34 NAVAL L. REV. 75, 78 (1985) (placing Albania and the Corfu Channel dispute in the East-West conflict context). 183. See Michael D. Nolan & Frederic G. Sourgens, Limits of Consent, Arbitration Without Privity and Beyond, in LIBER AMICORUM BERNARDO CREMADES 873 (Miguel Ángel Fernández-Ballesteros & David Arias, eds. 2010). 184. David S. Rubenstein, Immigration Blame, 87 FORDHAM L. REV. 125 passim (2018) (explaining the psychological mechanism in the immigration context). 185. Corfu Channel (U.K. v. Alb.), Judgment, 1949 I.C.J. 4, 32 (Apr. 9) (hinting that the document was about first shots being fired by Albania). 186. See Strong, supra note 2, at 137. 187. CARL SCHMITT, DER BEGRIFF DES POLITISCHEN 27 (1932). 132 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1

2. Witness Testimony Perhaps unsurprisingly, the specter of political decision is the more concrete when parties rely upon witness testimony to prove their case. This potentially undergirds the skepticism of many international jurists towards witness testimony as a good means to establish facts. This skepticism has already surfaced in the discussion of key international jurisprudence so far: the International Court of Justice found a way around witness testimony in Corfu Channel.188 Further, the Court appeared unwilling to make difficult credibility findings in the Pulp Mills case.189 This distant engagement with witness evidence is apparent throughout the Court’s jurisprudence.190 One of the chief reasons becomes apparent when examining the robust use of witnesses in a different international dispute resolution context: investor-state arbitration. As Corfu Channel has shown, witness testimony tends to ascribe motive (why did the vessels leave port?).191 This motive (or lack thereof) is frequently dispositive in the context of investor-state disputes.192 Thus, an issue may arise whether a government official purposefully delayed responding to a letter and therefore made it “disappear.” Witness testimony can establish when the document actually arrived and what happened to it next and why.193 Similarly, the question whether a government official was in fact authorized by his or her superiors to sign a key document—and why the official would have signed the document without authorization—is an issue as to which witness testimony can frequently be dispositive.194 Finally, the reason for increased governmental

188. Corfu Channel, 1949 I.C.J. at 17. 189. Pulp Mills, Judgment, 2010 I.C.J. 14, 72-73 (Apr. 20). 190. See Frontier Dispute (Benin v. Niger), Judgment, 2005 I.C.J. 90, 128 (July 12) (noting the lack of reliance by the party submitting testimony on the testimony in question). 191. Corfu Channel, 1949 I.C.J. at 17. 192. See Todd Weiler, Standards of Treatment, in OXFORD HANDBOOK ON INTERNATIONAL INVESTMENT LAW 259, 285 (Peter Muchlinski et al. eds., 2008) (noting the centrality of the concept of arbitrary conduct in investor-state arbitrations). 193. See, e.g., RSM Production Corp. v. Grenada, ICSID Case No. ARB/05/14, Award, ¶¶ 182-89 (Mar. 11, 2009). 194. See, e.g., Churchill Mining PLC and Planet Mining Pty Ltd v. Republic of Indonesia, ICSID Case No. ARB/12/14, Award, ¶ 306 (Dec. 6, 2016). 2020] TRUTHS IN TRANSLATION 133 action is frequently a matter that would not have been committed to writing.195 The importance of witness evidence led one investor-state tribunal to take the momentous step to compel a sitting minister to testify. In the Tulip v. Turkey arbitration, a key question was the reason why a housing project was terminated.196 Turkey submitted that it had commercial reasons for doing so.197 The claimant insisted that it had a right to examine the person with direct person knowledge of the underlying events—a person for whom Turkey had not submitted a witness statement.198 The person in question was a sitting minister in the Turkish government.199 The tribunal agreed and required that Turkey make the minister available for cross-examination.200 Such a result would be highly unusual in other contexts. The calling of a witness under these circumstances would likely increase the danger for international dispute resolution to slip into the kind of political spectacle that the Corfu Channel court so deftly managed to circumnavigate.201 It would therefore be unlikely that the witness would be called despite the fact that the witness’ testimony is clearly material for the outcome of the case. Witness testimony, in other words, is difficult for international justice precisely because it requires one to take sides. To take it seriously is to make findings of credibility that call into question politically laden narratives. Such decisions again bring the specter of blame and alternative facts to the heart of the international dispute resolution enterprise. International justice thus tends to treat witness testimony the way a groomer would a porcupine—with great care.

195. See, e.g., Apotex Holdings Inc. and Apotex Inc. v. U.S., ICSID Case No. ARB(AF)/12/1, Award, ¶ 3.30 (Aug. 25, 2014). 196. Tulip Real Estate Inv. & Dev. BV v. Turkey, ICSID Case No. ARB/11/28, Award, ¶ 60 (Mar. 10, 2014). The Author acted for Tulip Real Estate Investment & Development BV in the annulment proceeding in the same dispute. 197. Id. at ¶ 417. 198. Tulip Real Estate Inv. & Dev. BV v. Turkey, ICSID Case No. ARB/11/28, Annulment, ¶ 115 (Dec. 30, 2015). 199. Tulip Real Estate Inv. & Dev. BV v. Turkey, ICSID Case No. ARB/11/28, Award, ¶ 37 (Mar. 10, 2014). 200. Id. 201. See Section II.B. 134 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1

3. The Quest for “Corroboration” How does international justice seek to handle witness testimony with such care so as to avoid falling into the trap of political confrontation (as opposed to judicial dispute resolution)? The core principle used in international justice is to look for corroboration for potentially explosive witness evidence.202 The Court provided a general indication of the standard it would apply when examining the probative value of witness testimony: was the statement “relied on by the other Party” or “corroborated by impartial, neutral sources.”203 This quest for corroboration is an attempt to dislodge witness testimony from its place in a narrative and to replace it instead in the context of “neutral” facts.204 This type of corroboration can broadly come in three ways. First, a witness may simply provide greater detail or precision than what has already been established in broad outline by documentary evidence. In this case, the witness testimony would be corroborated by direct, documentary evidence. One such example of direct corroboration is the testimony of Professor Ackermann in the Temple of Preah Request for Interpretation case before the International Court of Justice.205 Professor Ackermann had visited the site at issue in the dispute in 1961 and provided a detailed discussion of his observations from that time.206 The Court made several findings of fact based on Professor Ackermann’s testimony.207 The Court did not make any direct credibility findings. Instead, the Court looked for corroboration for Professor Ackermann’s testimony. It found this corroboration in submissions by counsel, as well as in relevant maps. It thus relied upon the testimony together with “a number of other factors.”208 This treatment is consistent with other cases, in which the Court

202. Armed Activities on the Territory of the Congo (Dem. Rep. Congo v. Uganda), Judgment, 2005 I.C.J. Rep. 168, ¶ 298 (Dec. 19); Request for Interpretation of the Judgment of 15 June 1962 Concerning the Temple of Preah (Cambodia v. Thai.), 2013 I.C.J. 281, 312 (Nov. 11) (“Preah Interpretation”); Territorial and Maritime Dispute in the Caribbean Sea (Nicar. v. Hond.), Judgment, 2007 I.C.J. 659, 717-18 (Oct. 8). 203. Dem. Rep. Congo v. Uganda, 2005 I.C.J. Rep. at ¶ 298. 204. Id. 205. Preah Interpretation, 2013 I.C.J. at 312. 206. Id. 207. Id. at 312-14. 208. Id. at 313. 2020] TRUTHS IN TRANSLATION 135 appears to have relied in part upon witness testimony as part of its findings of facts.209 One of the features to note in this context is that the Court looks to documents to corroborate witness testimony. Whether this witness testimony is tested against licenses issued by the state proffering the witness or against maps, the Court tends to look to documents to confirm witness testimony and to use witness testimony where the documents already permit an inference as to the underlying facts in question—so long as the witness testimony can provide the Court with greater precision.210 Second, corroboration can be more attenuated. After all, much of the need for witness testimony in international disputes arises from the absence of documentary proof. One way to corroborate a witness account is to obtain further witness testimony to the same effect. Two different witnesses could corroborate each other’s accounts.211 Alternatively, if such witnesses are not willing or able to testify in person, contemporaneous reports may identify third- party statements consistent with the witness testimony.212 These reports would corroborate the witness through hearsay evidence.213 In both instances, the corroboration would come by way of direct or indirect witness testimony. Third, a witness may not be able to rely upon others for direct corroboration, at all. Rather, the corroboration will need to come from context.214 The witness in this instance would hold a puzzle piece that fits with the rest of the record evidence. Although there is no corroboration for the information on the witness’ piece of the puzzle, the fact that the witness’ testimony fits with the rest of the picture already assembled by the record can also provide a

209. Territorial and Maritime Dispute in the Caribbean Sea (Nicar. v. Hond.), Judgment, 2007 I.C.J. 659, ¶ 190-92, 195(Oct. 8). 210. See id.; Preah Interpretation, 2013 I.C.J. at ¶ 86. 211. See Churchill Mining PLC v. Indon., ICSID Case No. ARB/12/14, Award, ¶ 141 (Dec. 6, 2016). 212. See SOURGENS ET AL., supra note 87, at 292 (discussing circumstantial hearsay evidence). 213. Id. 214. See Salini Costruttori S.P.A. v. Morocco, ICSID Case No. ARB/00/4, Decision on Jurisdiction, ¶ 34 (July 16, 2001) (using the legal framework established independently by the tribunal to corroborate its understanding of the nature of a contract). 136 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 different, circumstantial kind of corroboration for the witness’ account.215 This quest for corroboration can play out in unexpected ways. The Tulip v. Turkey arbitration again is instructive.216 As discussed in the previous section, the Tulip tribunal took the momentous step to require live testimony from a sitting minister to shed light onto the state’s motives for ultimately terminating the investment relationship with Tulip.217 The Tulip tribunal took the less momentous step to ignore the minister’s testimony—despite the fact that it broadly supported Tulip’s case and tended to detract from Turkey’s defense.218 The Tulip tribunal did not explain its decision not to mention the minister’s testimony.219 An annulment panel asked to review the award did: the tribunal’s choice was “an illustration of exercising its judicial function of choosing which evidence it finds relevant and which it does not.”220 The tribunal found that the conclusion drawn from the minister’s testimony by Tulip was “not established by the evidentiary record.”221 In other words, the tribunal did not find corroboration from other sources (i.e. “the evidentiary record”) for the witness’ statement and proceeded to ignore the testimony consistent with its “judicial function.”222 The conclusion may be unexpected as a matter of material justice—how is it consistent with a tribunal’s justice to call a witness if not to address his testimony? It is however sadly consistent with, and illustrates the price of, the corroboration frame in international jurisprudence.223 Given this potential for unexpected outcomes, it is important to understand what precisely international courts and tribunals do when they look for corroboration. All three types of corroboration ultimately depend upon different ways in which a tribunal can draw an inference. The use of inferences is most visible when there

215. See Churchill Mining PLC, ICSID Case No. ARB/12/14, at ¶ 140. 216. Tulip Real Estate Inv. & Dev. B.V. v. Turk., ICSID Case No. ARB/11/28, Award, ¶ 60 (Mar. 10, 2014). 217. Id. 218. Tulip Real Estate Inv. & Dev. B.V. v. Turk., ICSID Case No. ARB/11/28, Decision on Annulment, ¶¶ 154-55 (Dec. 30, 2015). 219. Id. 220. Id. at ¶ 149. 221. Id. 222. Tulip Real Estate Inv, ICSID Case No. ARB/11/28 at ¶ 149. 223. In the interest of disclosure, the Author served as an expert for Tulip in the annulment proceedings but has no current relationship with the parties. 2020] TRUTHS IN TRANSLATION 137 is no direct evidence that would support the witness’ testimony but the witness testimony is consistent with other facts that have been independently established. Corroboration and inference here work visibly hand in glove: we confirm the testimony (the missing puzzle piece) because we infer from context (the puzzle we have already put together) that it “fits” and is accurate. The use of inferences is similarly present in the other two scenarios. The presence of documentary proof that is broadly consistent with the testimony, if less precise than the testimony, corroborates the witness testimony because we infer from the general documentary proof to the specific witness testimony on point. This inference functions like a forensic picture analysis on a television crime drama. Those dramas frequently create cleaned up, zoomed images from a blurry native file. We (and the TV detectives) assume that the cleaned up, zoomed image is in fact a feature of the reality pictured in the native file rather than an external imprint left upon the image by the zooming process.224 This assumption is an inference from the similarity between the enhanced and native image. Finally, the corroboration of one eyewitness’ account of an event with another eyewitness’ testimony also relies upon proof by inference. Multiple people identify the defendants in My Cousin Vinny as the robbers of the Sac-o-Suds convenience store.225 These layers of consistent testimony create only an inference that the perception of each of the witnesses was accurate and credible. However, this does not prove directly that each witness reliably saw what they say they saw. In fact, as My Cousin Vinny illustrates, the witnesses were all mistaken.226 The witnesses corroborated each other’s story but did not prove that the testimony was in fact true by more than an inference that two people see more or see better than one person. To corroborate something means to support an inference as to its truthfulness, but inferences are not without problems in their

224. See Sebastian Anthony, CSI-Style Super-Resolution Image Enlargement? Yeeaaah!, EXTREMETECH.COM (July 18, 2012), http://www.extremetech.com/extreme/132950-csi-style-super-resolution-image- enlargment-yeeaaaah [https://perma.cc/2H35-X2LY] (discussing the technology used in crime shows and its achievability in real life). 225. MY COUSIN VINNY (20th Century Fox 1992). 226. See id. (Vinny Gambini: “Well, perhaps the laws of physics cease to exist on your stove. Were these magic grits?”). 138 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 own right. They can cause a finder of fact to ignore evidence that is uncorroborated.227 As My Cousin Vinny shows, they could also cause a finder of fact to make findings that do not reflect the true probative value of the evidence.228 Thus, they are not the complete answer to the question how we find accurate information about what happened in international justice. Rather, they raise the next question. To understand our desire to corroborate accounts— shared by international jurists and viewers of crime dramas alike—is to understand reasoning by inference. The move away from stepping into the overtly politically laden confrontation of blame, narrative, and alternative fact thus comes at a potentially considerable cost. To appraise this cost—and potentials for correction—is critically dependent upon a closer inspection of proof by inference. So how does it work?

B. Inferences Inferences are at the heart of proving facts in international justice. Documents hardly ever tell the whole story. One could try to close this gap with witness testimony, but international courts and tribunals typically require corroboration for witness testimony.229 This corroboration is based on some form of inference that the direct witness evidence is credible and probative.230 As already set out above, this means that international justice uses corroboration—and thus inferences—as the main tool to escape from the dilemma of political confrontation and its descent into “alternative facts.” However, the problem goes deeper than that. In many cases, parties are unlikely to find witness evidence of the critical events at the heart of their submissions. In those instances, a party will have to argue for an inference from the record evidence they could collect that a different event also must have taken place. The Corfu Channel dispute is a good illustration of this larger problem. The UK could not find a witness of the mine-laying for obvious reasons of military secrecy (or if it could find one, the UK

227. See Tulip Real Estate Inv. & Dev. BV, ICSID Case No. ARB/11/28, at ¶ 149. 228. For a detailed account of the psychological mechanisms underlying overreliance on eyewitness testimony see Heller, supra note 122, at 248. 229. See Section III.A.3, supra. 230. See Section III.A.3, supra. 2020] TRUTHS IN TRANSLATION 139 certainly did not proffer them).231 Instead, it submitted testimony from a defector who witnessed the loading of mines onto Yugoslav vessels and had access to the schedule upon which the vessels left port and returned to port.232 On this basis, the UK asked for a finding that Albania had permitted the Yugoslav navy to mine the North Corfu straight.233 This finding would have been an inference. This inference did not corroborate a witness’ testimony. Rather, it asked for a finding of a fact for which there was no direct evidence in the record, at all.

1. Inferences as Gap Fillers Outside of the context of corroboration, inferences fill gaps. Here, inferences work like the puzzle pieces discussed in the context of corroboration. Suppose the record evidence provides an incomplete picture of events. Rather than relying upon witness testimony to fill in the gaps—and to fill them in a manner that the rest of the evidence would corroborate—suppose the parties do not submit further evidence as to what occurred. In this context, a finder of fact would have two alternatives. It could make a decision based upon burdens of proof. The party that failed to provide some direct evidence would lose the case precisely because there is a gap in the puzzle. Alternatively, the finder of fact could permit the parties to make submissions of how the finder of fact should fill out the remainder of the puzzle. In the second scenario, the finder of fact would permit the parties to prove facts by inference. As the discussion so far has already shown, international justice has chosen the second path and permits proof by inference. International jurisprudence follows a five factor test in proving facts by inference.234 In the first instance, the finder of fact will establish the likelihood that the fact could have been proved by direct evidence.235 If a party should have been able to prove a fact by direct evidence but chose not to do so, finders of fact are far more reluctant to make such a finding.236 At a minimum, finders of

231. Corfu Channel, 1949 I.C.J. at 8. 232. Id. at 12-17. 233. Id. 234. SOURGENS ET AL., supra note 87, at 295 (Evidentiary Principle § 28). 235. Id. 236. Id. 140 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 fact will require an express or implicit explanation as to why the party chose not to put forward direct evidence in support of its case.237 Next, it is important to assess the probative value of the direct evidence that has been proffered requesting an inference.238 Is the fact on the basis of which the party requests an inference proved? The firmer the footing of the facts surrounding the empty space on the record, the more likely it is that these facts will be able to bear the additional weight of an inference. Thus, in Corfu Channel, the Court may have wondered whether LCdr Kovacic was a credible witness and in fact saw ships loading mines etc.239 The inference would only ever be warranted if the Court had believed the loading of mines to have taken place. Further, the length of the leap from the evidence to the inference to be drawn needs to be assessed.240 If one puts together a puzzle of Mickey Mouse, and all that is missing is a three-piece cluster and Mickey Mouse only has one ear, it is not a significant leap from the surrounding evidence to conclude that the missing pieces pictured Mickey’s ear. Guessing what a missing three-piece cluster of a Jackson Pollock puzzle would show may be significantly more challenging.241 The closer the link between the inference and the evidence, the more likely it is that a court or tribunal would make the relevant finding of fact. It is also important to consider how many pieces of evidence independently support the drawing of the inference in question.242 Inference, in other words, can be corroborated by multiple pieces of independent evidence. In the context of the puzzle analogy, the more pieces of the puzzle one has assembled, the more likely one can hazard a guess as to what is pictured in the empty space. Each puzzle piece laid is additional support for the inference.

237. Id. 238. Id. 239. Corfu Channel, Judgment, 1949 I.C.J. 4, 16-17 (Apr. 9). 240. SOURGENS ET AL., supra note 87, at 295. 241. See Roberta Smith, Drips, Dropped: Pollock and His Impact, N.Y. TIMES (Dec. 31, 2015), https://www.nytimes.com/2016/01/01/arts/design/review-drips-dropped- pollock-and-his-impact.html [https://perma.cc/B225-S4RV] (discussing Pollock’s abstract expressionism and its impact). 242. SOURGENS ET AL., supra note 87, at 295. 2020] TRUTHS IN TRANSLATION 141

A final factor is how significant the inference is to the case.243 The more significant an inference, the more care a tribunal will give to establishing the basis for drawing it.244 The less significant an inference, the more willing a tribunal will be to draw inferences from a relatively bare record. These factors in drawing inferences are not elements. Rather, they follow a typical factor test.245 They give the tribunal significant discretion in approaching its fact-finding function.246 They also mutually affect each other, meaning that it is not possible to isolate the decision to draw or not to draw any one inference to exclusively one factor.

2. The Threshold Question of Reasonableness In drawing inferences, a tribunal first must answer a threshold question: is the inference requested by a party reasonable on its face? Classically, international law scholarship submits that a court or tribunal draws an inference on the basis of such reasonableness only.247 It is easy to imagine an inference that would on the whole be unreasonable. For example, Yugoslavia had a navy in 1946.248 But it would be unreasonable to conclude on the basis of this fact alone that Yugoslavia used its navy to lay mines in the North Corfu Straight in 1946. The submission has no further factual anchor and resembles conspiracy theory and speculation rather than proof of that speculation. Neither Yugoslavia nor Albania would have had to rebut such evidence from the UK in the Corfu Channel case. It would have been insufficient even to suggest the inference on its face. There is a minimum floor of concreteness to draw an

243. Id. 244. Corfu Channel, 1949 I.C.J. at 17. See also Case Concerning Oil Platforms (Islamic Republic of Iran v. U.S.), Judgment 2003 I.C.J. 225, 234-35 (Sep. Op. Higgins). 245. SOURGENS ET AL., supra note 87, at 295. 246. See Alapli Elektrik B.V. v. Turk., ICSID Case No. ARB/08/13, Decision on Annulment, ¶ 234 (July 10, 2014) (“What is more, a tribunal has considerable discretion in its evaluation of the evidence.”). 247. MOJTABA KAZAZI, BURDEN OF PROOF AND RELATED ISSUES: A STUDY ON EVIDENCE BEFORE INTERNATIONAL TRIBUNALS 259 (1996); SANDIFER, supra note 69, at 173. 248. Yugoslav Navy: Organization, Personnel, Installation, Ordnance, CENT. INTELLIGENCE AGENCY 1, 1 (Aug. 2, 1963), https://www.cia.gov/library/readingroom/docs/CIA-RDP80T00246A069000460001- 5.pdf [https://perma.cc/8YFW-58XW]. 142 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 inference. International law deals with this floor in terms of reasonableness.249 Although it is difficult to quantify the floor needed for the drawing of a reasonable inference, the difference is on the whole intuitive. It compares the inference to be drawn to our general expectation of how the world (ought to) behave. It refers to basic presumptions about how we expect the world to operate much like the inference in Corfu Channel that it is reasonable for Albania to have observed its strategic coast line, particularly when it observed it diligently on other occasions thus closing the gap between the inference and record fact considerably.250 In short, it requires that in the eyes of the tribunal, a party satisfies each of the factors listed in the previous section. If there were no other rebuttal evidence, could a court or tribunal draw an inference in favor of the moving party without failing to provide adequate reasons for its decision? This reasonableness assessment does not operate in a vacuum. Rather, it uses our general expectations of how the world works, all things being equal. It measures the inference requested against this general background understanding. If we ask a person what the three missing puzzle cluster of the picture of our one- eared Mickey Mouse is likely to show, we tacitly assume that the person has a seen a picture of Mickey Mouse before. As will be discussed later on, this assessment thus makes necessary an engagement with presumptions when making an initial reasonableness assessment.

3. Plausibility—The Choice Between Reasonable Alternatives In many scenarios, the problem is that there is more than one reasonable inference one can draw from the evidence. To give an everyday example, a parent is in the kitchen while a younger brother, Michael, and his older sister, Josephine, are playing in the den a room over. They are playing a game that involves a fight between Lego figures. There is a thump. The younger brother cries. The parent walks over and finds Michael sprawled crying on the floor holding his leg. His sister Josephine stands over him, Lego

249. KAZAZI, supra note 247; SANDIFER, supra note 69. 250. Corfu Channel (U.K. v. Alb.), Judgment, 1949 I.C.J. 4, 18-20 (Apr. 9).

2020] TRUTHS IN TRANSLATION 143 figure in hand, shouting “bad Garmadon!” Did Josephine kick or push Michael? The inference would be reasonable in light of the typical brother-sister dynamics in a heated exchange that developed out of a game. Or did Michael trip and fall? The inference would also be reasonable in light of children’s occasional loss of spatial awareness while playing with Legos. For the parent to establish what happened, the parent will have to compare both these reasonable inferences to each other. The comparison of reasonable inferences is rarely made express in international jurisprudence. It can however be observed reasonably cleanly in the Croatian Genocide case.251 The International Court of Justice there dealt with the question whether Serbia had the requisite specific intent to commit genocide of the Croatian population in the civil war following the in the 1990s.252 Croatia submitted that this intent could be inferred from the overall record, listing 17 specific factual circumstances which supported the inference.253 Serbia argued that it lacked specific genocidal intent towards the Croatian population at issue in the case.254 The Court suggested that it would draw an inference of specific genocidal intent if that intent was the only reasonable inference from the record.255 The Court then examined the factual circumstances submitted by Croatia. In examining these factual circumstances, the Court found that in some instances, Serbian forces had evacuated Croatian civilians from the heaviest scene of fighting.256 This evacuation together with other similar factual elements, the Court found, meant that genocidal intent was not the only reasonable inference from the evidence before the Court.257 It is tempting to take the Court at its word: Croatia had not overcome the high hurdle of disproving that an alternative inference from the record evidence was also reasonable. This conclusion does not fully do justice to what the Court found. The Court’s analysis ultimately did not suggest that Serbia may have

251. See Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Croat. v. Serb.), Judgment, 2015 I.C.J. Rep. 3, ¶ 119 (Feb. 3). 252. Id. 253. Id. at 119-20. 254. Id. 255. Id. at 128. 256. Id. at 127. 257. Id. at 127-28. 144 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 acted with genocidal intent but that Croatia had failed to submit sufficient evidence to rebut a reasonable inference that it had not.258 Rather, the Court’s focus on the evacuation of the Croatian population in particular suggests that the Court made a stronger finding in Serbia’s favor: on the record before the Court, the Court inferred that Serbia lacked the relevant intent.259 Following this chain of reasoning, the Court examined the potential inferences it could draw against each other.260 It then determined that one inference was more readily supported by the evidence taken as a whole compared to the other.261 It thus concluded that the plausible inference from the record was that Serbia lacked the intent in question despite the fact that Croatia’s submission of genocidal intent passed the initial reasonableness screen for the drawing of inferences.262 This comparative approach to the drawing of inferences is consistent with international arbitral jurisprudence, as well. That jurisprudence is frequently asked to make determinations of facts based on inferences when the tribunal is faced with multiple reasonable proffered inferences from the parties.263 The jurisprudence compares the strength of the inferences against each other along.264 This jurisprudence, too, suggests that international jurisprudence relies upon a plausibility analysis in

258. Croat. v. Serb., 2015 I.C.J. 119-28. 259. Id. 260. Id. 261. Id. 262. Id. 263. See, e.g., Rompetrol Grp. N.V. v. Rom., ICSID Case No. ARB/06/3, Award, ¶ 232 (May 6, 2013); Cambodia Power Co. v. Cambodia, ICSID Case No. ARB/09/18, Award, ¶¶ 82–83, 94–95 (Mar. 22, 2011); Siag v. Egypt, ICSID Case No. ARB/05/15, Award, ¶ 12 (June 1, 2009); Tokios Tokelès v. Ukr., ICSID Case No. ARB/02/18, Award, ¶ 3 (June 29, 2007); Methanex Corp. v. U.S., UNCITRAL–NAFTA, Award, ¶¶ III.52–III.57 (Aug. 3, 2005); CDC Grp. Pub. Ltd. v. Republic of Seychelles, ICSID Case No. ARB/02/14, Award, ¶¶ 45–51 (Dec. 17, 2003); Feldman v. Mex., ICSID Case No. ARB(AF)/99/1, Award, ¶¶ 32–37 (Dec. 3, 2002) (dissenting opinion of Jorge Covarrubias Bravo). 264. See, e.g., Rompetrol Grp. N.V. v. Rom., ICSID Case No. ARB/06/3, Award, ¶ 232 (May 6, 2013); Cambodia Power Co. v. Cambodia, ICSID Case No. ARB/09/18, Award, ¶¶ 82–83, 94–95 (Mar. 22, 2011); Siag v. Egypt, ICSID Case No. ARB/05/15, Award, ¶ 12 (June 1, 2009); Tokios Tokelès v. Ukr., ICSID Case No. ARB/02/18, Award, ¶ 3 (June 29, 2007); Methanex Corp. v. U.S., UNCITRAL–NAFTA, Award, ¶¶ III.52–III.57 (Aug. 3, 2005); CDC Grp. Pub. Ltd. v. Republic of Sey., ICSID Case No. ARB/02/14, Award, ¶¶ 45–51 (Dec. 17, 2003); Feldman v. Mex., ICSID Case No. ARB(AF)/99/1, Award, ¶¶ 32–37 (Dec. 3, 2002) (dissenting opinion of Jorge Covarrubias Bravo). 2020] TRUTHS IN TRANSLATION 145 finding facts when faced with rival submissions regarding what inferences the record supports.265 This plausibility analysis in one sense greatly advances the quality of justice available before international courts and tribunals. International courts and tribunals seek to approximate to the best of their ability what happened on the ground.266 They do not rely upon overly technical rules that would create greater risks of mischaracterizing events. Inferences thus do not on their face seek to distort fact finding in favor of the moving or the non- moving party. But the plausibility analysis highlights the blind spot of inferences already discussed in the previous section. The comparison of different inferences implicitly relies upon an understanding of how the world operates, all things being equal, to ascertain which inference is more probable in light of the record evidence.267 This general understanding of how the world operates already informs the threshold question whether a proffered inference would reasonably be entailed by the record. This analysis is not one of seeking to establish that the elements that an inference is appropriate have been met. Rather, it is one that engages the sound discretion and judgment of the finder of fact. This component of judgment or discretion is further heightened when a finder of fact is asked to compare how much more reasonable one inference is when compared to another. This exercise relies even more heavily upon the background assumptions as to which inference deviates further from some hypothetical default condition of everyday life. This background condition informs how we can distinguish between reasonable inferences when there is no discernible quantitative difference between rival inferences along the test outlined in the previous section. A plausibility analysis thus is only as strong as the presumptions upon which it relies.

265. SOURGENS ET AL., supra note 87, at 150-53. 266. Feldman, ICSID Case No. ARB(AF)/99/1 at ¶ 36 (dissenting opinion of Jorge Covarrubias Bravo). 267. SOURGENS ET AL., supra note 87, at 152. 146 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1

C. Presumptions If inferences are at the heart of fact finding in global justice, presumptions are like the sinus node causing the heart to beat according to a certain rhythm.268 Presumptions are deeply embedded in the drawing of record inferences.269 At the same time, presumptions themselves are a certain kind of non-record inference that sets the conditions against which a court or tribunal can make finding of facts from the record.270 As will be discussed in detail in the next sections, presumptions establish general expectations about how the world operates. These general expectations about how the world operates are not of themselves probative of anything. They become probative only if and when we apply them to a specific dispute. Then, we compare how the metaphorical puzzle pieces of a particular record measure up against our general expectation of how the puzzle should be put together or what the final puzzle should look like. Presumptions thus constantly project over the unchartered territory of the record evidence and help put it together into a map for resolving the dispute. As esoteric as the function of presumption may sound, presumptions are nothing new or mysterious in the international law of evidence. Rather, global dispute resolution has a robust doctrinal framework for dealing with presumptions.271 This doctrinal framework breaks presumptions into two different kinds, legal presumptions and judicial presumptions.272 Before delving more deeply into presumptions in their own right, this section briefly introduces the doctrinal understanding of these two kinds of presumptions.

268. Moinuddin Choudhury et al., Biology of the Sinus Node and its Disease, 4 ARRHYTHM ELECTROPHYSIOLOGY REV. 28, 28 (2015). 269. See supra Section III.B (inferences); infra Section III.C.2 (application to presumptions). 270. SOURGENS ET AL., supra note 87, at 117. 271. Id. at 111-34; KOLB, supra note 69, at 241-43; SANDIFER, supra note 69, at 142; Principle XII.3, TRANS-LEX.ORG, https://www.trans-lex.org/968000 (last visited Sept. 10, 2020). 272. SOURGENS ET AL., supra note 87, at 111-34; KOLB, supra note 69, at 241-43. 2020] TRUTHS IN TRANSLATION 147

1. Legal Presumptions as Part of Applicable Law Legal presumptions are the most straightforward way in which presumption overlay the fact-finding process. Legal presumptions are part of the applicable law.273 A legal presumption is a rule of applicable law that requires a decisionmaker to use a certain rubric when approaching the record.274 Due to its television fame, one of the most commonly known legal presumption is the presumption of innocence in criminal proceedings.275 The presumption of innocence in the US context is a rule of evidence or proof founded in the constitutional principle of due process.276 It requires that the finder of fact in a criminal trial assess the evidence with the assumption of the accused’s innocence.277 The presumption of innocence thus applies by constitutional mandate to US criminal trials, in theory, at least.278 To ignore it would be to violate the defendant’s constitutional rights in a way analogous to permitting evidence to be introduced against a criminal defendant in violation of the Confrontation Clause of the US Constitution.279 Although the issue has been debated, general international law recognizes certain kinds of legal presumptions.280 For instance, general international law presumes that state law and specialized international law is consistent with general international law.281 General international law also presumes that state conduct such as

273. SOURGENS ET AL., supra note 87, at 111-34; KOLB, supra note 69, at 241-43. 274. SOURGENS ET AL., supra note 87, at 293. 275. David G. Post, Nelson v. Colorado: New Life for an Old Idea?, CATO SUP. CT. REV., 2016-2017, at 205, 215. 276. See Shima Baradaran, Restoring the Presumption of Innocence, 72 OHIO ST. L.J. 723 passim (2011) (arguing for a more robust due process reading of the presumption). 277. Post, supra note 275, at 217. 278. See Baradaran, supra note 276, at 754-65 (discussing the limitations of the presumption in the criminal procedure setting and its roots in constitutional misunderstanding). 279. Pamela R. Metzger, Confrontation as a Rule of Production, 24 WM. & MARY BILL RTS. J. 995, 1037 (2016); François Quintard-Morénas, The Presumption of Innocence in the French and Anglo-American Legal Traditions, 58 AM. J. COMP. L. 107, 140-41 (2010). 280. See ANNA RIDDELL & BRANDAN PLANT, EVIDENCE BEFORE THE INTERNATIONAL COURT OF JUSTICE 102 (2009) (questioning existence of legal presumptions). But see Kolb, supra note 69, at 944 (confirming the presence of legal presumptions in general international law). 281. HEEJIN KIM, REGIME ACCOMMODATION IN INTERNATIONAL LAW: HUMAN RIGHTS IN INTERNATIONAL ECONOMIC LAW AND POLICY 35 (2016). 148 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 the passing of legislation or the use of executive power is valid.282 These presumptions organize the manner in which an international law fact finder will interpret the state law or governmental decisions in question.283 However, it radiates more broadly in how the international finder of fact organizes other record facts that are dependent upon the law or decision—that is, how it will organize its appraisal of the implementation of law or the carrying out of a decision in a specific dispute by requiring that the pieces of the factual puzzle be put together on the assumption that the underlying legislation comported with international law and was domestically valid. As a default matter, legal presumptions are rebuttable.284 A party can introduce evidence to show that the presumption is inapplicable in the case at bar.285 This means that factual findings do not always follow the default rules of legal presumptions.286 It also means that the rubric through which a finder of fact engages the record remains flexible even in the face of legal presumptions. However, the existence of legal presumptions creates a significant obstacle for any party wishing for a court or tribunal to depart from them. As the presumption of innocence shows, this obstacle is not fatal—plenty of criminal defendants are ultimately convicted,287 but as the presumption of innocence also shows, it is a significant obstacle nonetheless.288

2. Judicial Presumptions Based on Relevant Practice Judicial presumptions are less straight forward than legal presumptions. Unlike legal presumptions, they are not the result of the application of specific legal mandate from the applicable law.

282. CHENG, supra note 12, at 305. 283. KIM, supra note 281, at 35; CHENG, supra note 12, at 305. 284. SOURGENS ET AL., supra note 87, at 123-24. 285. Id. 286. Id. 287. What is the Probability of Conviction for Criminal Defendants?, BUREAU OF JUSTICE STATISTICS, https://www.bjs.gov/index.cfm?iid=403&ty=qa [https://perma.cc/K6XS- MYYV] (last visited Sept. 9, 2020) (summarizing US federal conviction rates). 288. See id.; see also James Q. Whitman, Presumption of Innocence or Presumption of Mercy?: Weighing Two Western Modes of Justice, 94 TEX. L. REV. 933, passim (2016) (discussing the function of the presumption of innocence from a comparative perspective). 2020] TRUTHS IN TRANSLATION 149

Instead, they apply the general inductive principle that one should expect the record to follow the pattern of general practice.289 One of the most mundane codifications of judicial presumptions is the US Federal Rule of Evidence governing habit evidence.290 The rule permits a party to plead that on a particular occasion, a person acted consistently with a habit despite the fact that there is no evidence whatsoever to corroborate that the person acted consistently with their habit on the particular occasion.291 The rule requires proof of the habit itself.292 A party satisfies this requirement when it can show that a person has a frequent, consistent response to the same specific stimulus.293 The point of proof of a fact by habit is that one can substitute the proof of a specific contested fact (“A did B in response to C on day X”) with proof of a different factual predicate: A did B in response to C on all days D to W on which A was exposed to C.294 Although we do not have proof of A’s action on day X, we can substitute the consistency of practice from other contexts to stand in for an unknown. International law permits a similar kind of proof by means of judicial presumptions.295 Judicial presumptions require proof of a general pattern of conduct by like-situated parties akin to a usage of trade.296 It must then place the party to whom the conduct will be attributed within that general pattern.297 Once that has been done, the proof of the general conduct or practice will create a presumption that the person in question acted consistently with the general pattern or practice on the occasion in question even if there is no other proof to that effect.298 The effect of this judicial presumption is that the record is appraised consistently with such habit. In some instances, there

289. SOURGENS ET AL., supra note 87, at 117-21. 290. FED R. EVID. 406. 291. Id. 292. Id. 293. Barrett J. Anderson, Recognizing Character: A New Perspective on Character Evidence, 121 YALE L.J. 1912, 1916 n.10 (2012). 294. See Reyes v. Mo. Pac. R. Co., 589 F.2d 791, 795 (5th Cir. 1979) (setting out the regularity requirement in FED. R. EVID. 406). 295. See ROBERT KOLB, THE ELGAR COMPANION TO THE INTERNATIONAL COURT OF JUSTICE 241 (2014) (presumptions of facts or praesumptiones hominis). 296. Id. 297. Id. 298. Id. 150 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 will be genuinely no proof concerning a specific fact. At that point, the judicial presumption steps out into the open. The Burlington Resources Inc. v. Ecuador arbitration presents an example of such a use of a judicial presumption.299 The case arose out of an oil and gas project in Ecuador.300 The project involved an operator in charge of the day-to-day operations of the oil and gas project as well as “non-operators,” who contribute money and know-how to the project but do not control day-to-day operations.301 One of the contested issues in the arbitration was whether the non-operators had communicated with the government in a satisfactory manner.302 The non-operators relied upon a letter sent by the operator to the government.303 They proved up that as a trade usage in the oil and gas industry, the operator communicates on behalf of all project participants.304 In this case, there was no other evidence whether the operator was speaking on behalf of itself alone in this instance or for the project as a whole.305 The presumption therefore provided a fact by proof of the broader industry practice and proof that the project participants fit in that practice. However, as with legal presumptions, the judicial presumption also operates in the background to organize record material. It provides the rubric against which inferences are tested and puzzle pieces put together. Judicial presumptions, in other words, are a powerful organizing tool for fact finding even when they are not directly employed. The Kim v. Uzbekistan arbitration is an example of how judicial presumptions can fulfill such a role— and be ultimately dispositive of a core issue in the case.306 Relevantly, Uzbekistan submitted in the arbitration that the

299. Burlington Resources Inc. v. Ecuador, ICSID Case No. ARB/08/5, Decision on Jurisdiction, ¶¶ 326-29 (June 2, 2010). 300. Id. 301. See id.; see also DAVID E. PIERCE, TRANSACTIONAL EVOLUTION OF OPERATING AGREEMENTS IN THE OIL AND GAS INDUSTRY 16 (2008) (discussing operating agreements in the oil and gas industry). 302. See Burlington Resources Inc., ICSID Case No. ARB/08/5 at ¶¶ 326–329. 303. Id. 304. Id. 305. Id. (discussing the evidence of the letter and earlier proceedings in which Respondent appeared to treat the operator as speaking for the project rather than just itself as evidence of this general practice applying in the case). 306. Kim v. Republic of Uzb., ICSID Case No. ARB/13/6, Decision on Jurisdiction (Mar. 8, 2017). 2020] TRUTHS IN TRANSLATION 151 tribunal lacked jurisdiction over the dispute because the claimants had acquired their investment through corruption.307 Proof of corruption would indeed have divested the tribunal of jurisdiction under the relevant consent instrument.308 The case ultimately boiled down to a fight between judicial presumptions. Uzbekistan lacked direct evidence of corruption.309 Uzbekistan instead relied upon red flags in the investment structure.310 Particularly, Uzbekistan submitted that convoluted and complex aspects of the acquisition of the investment by claimants raised such red flags.311 To substantiate why this structuring raised red flags, Uzbekistan looked to indicia of corruption in general arbitral jurisprudence.312 The claimants, on the other hand, argued that “[t]he presence of many corporate layers are elements commonly seen in transactions in the CIS region.”313 In essence, claimants made an argument that their investment structuring should benefit from “green flags” because their structuring mirrored that of industry practice in the region.314 The tribunal ultimately resolved the case against Uzbekistan because Uzbekistan had failed to establish “the link between the advantage bestowed and the improper advantage obtained.”315 It went on to state “[t]he casting of doubt or aspersions as to the probity of a transaction is not sufficient.”316 In effect, the evidence of red flags and conduct consistent with the red flags was merely a “casting of doubt or aspersions.”317 This insufficiency was not purely that there was no direct proof of a quid quo pro (this would hardly ever be possible).318 The question rather was whether the

307. Id. at ¶ 543. 308. Id. 309. Id. at ¶ 548. 310. See id. at ¶ 546. 311. Id. 312. Id. 313. Kim, ICSID Case No. ARB/13/6, ¶ 547. 314. Id. 315. Id. at ¶ 589 (quoting Sistem Mühendislik İnşaat Sanayi ve Ticaret A.Ş. v. Kyrgyz Republic, ICSID Case No. ARB (AF)/06/1, Judgment, ¶ 43 (Sept. 9, 2009). 316. Kim, ICSID Case No. ARB/13/6 at ¶ 589. 317. Id. 318. See ALOYSISUS LLAMZON, CORRUPTION IN INTERNATIONAL INVESTMENT ARBITRATION 194-98 (2014) (discussing the jurisdictional jurisprudence addressing corruption allegations). 152 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 tribunal accepted the alternative explanation for the record evidence as consistent with the general business practice pled by the claimants.319 In other words, the Kim claimants won the judicial presumption battle. The tribunal examined the evidence through the lens of, and organized it consistent with the “green flags” submitted by claimants, not the “red flags” submitted by Uzbekistan. As with legal presumptions, judicial presumptions are rebuttable.320 A party can introduce evidence to show that the presumption is inapplicable in the case at bar.321 This means that factual findings do not always follow the default rules of judicial presumptions, either.322 Despite this, the obstacle created by judicial presumptions in some respects can be harder to overcome than legal presumptions. The stronger the evidence of a judicial presumption the more the finder of fact will be locked in to a path dependent perspective in appraising the record.323 Path dependence is a difficult thing to overcome.324 It is precisely for this reason that much of the work of advocates is to convince the finder of fact that a dispute at bar fits into a pattern for which the finder of fact has a ready rubric to hand.

IV. PRESUMPTIONS’ BIAS PROBLEM The discussion in the previous three Parts of this Article demonstrate that presumptions are central to the fact-finding process. Presumptions supply facts where the record otherwise would not yield them and thus avoids or modifies the otherwise strict applications of burdens of proof. Further, presumptions also organize the manner in which we must approach the necessarily incomplete pieces of a factual record to put the puzzle of the case back together. The power of presumptions is that they allow a decisionmaker to attempt to do material justice even in the context of imperfect information. Given the stakes in international justice,

319. See id. 320. SOURGENS ET AL., supra note 87, at 120-21. 321. Id. 322. Id. 323. See Oona Hathaway, Path Dependence in the Law: The Course and Pattern of Legal Change in a Common Law System, 86 IOWA L. REV. 601, 626-27 (2001) (discussing path dependence). 324. Id. at 643-44 (discussing how rarely path dependence can be overcome). 2020] TRUTHS IN TRANSLATION 153 the need to make decisions in the face of imperfect information is by no means an accident. As the Corfu Channel case showcased, it is an inherent feature of the sensitive questions that international judges and arbitrators are asked to resolve.325 However, the power of a presumption is also its Achilles heel. Presumptions operate as more or less articulated assumptions. These assumptions establish facts central to international disputes on the basis of what ought to be the case. Presumptions in other words encounter a classic moral problem in reverse: moral philosophy since the time of 18th century Scotsman, enlightenment bugbear, and common-sense apostle, David Hume, had an is-ought problem “that an ‘ought’ cannot be derived from an ‘is.’”326 Presumptions seek to infer an is (what actually happened) from an ought (what was supposed to have happened) and thus encounter the same inferential problem. This reverse is-ought problem reveals latent powerful biases in the international law of evidence that must be addressed lest the international rule of law falls victim to creating its own “alternative facts” rather than seeking to establish a more objective basis for decision. To put it bluntly, what seemed like the potential antidote to anxiety mongering in post-truth discourse now appears to be headed into precisely the same direction as post-truth discourse.327 Rule of law, too, mandates that facts be established within the frame of Rules for Radicals’ “experience of your own people.”328 To unpack this statement, rule of law has a “people” whose experience becomes paramount: its liberal champions.329 Liberal champions of rule of law thus become a competitor for and in policy space rather than fair regulators of that space.330 Rule of law itself becomes political.331 Populist leaders from Vladimir Putin

325. Corfu Channel, Judgment, 1949 I.C.J. Rep. 4, 16-18 (Apr. 9). 326. Geoff Sayre-McCord, Metaethics, STAN. ENCYCLOPEDIA PHIL. (Jan. 26, 2012), https://plato.stanford.edu/entries/metaethics/#IsOOpeQueArg [https://perma.cc/W4GY-ZHBE]. 327. REISMAN, supra note 131, at 67. 328. ALINSKY, supra note 128, at 127 329. See Brian Z. Tamanaha, The Dark Side of the Relationship Between the Rule of Law and , 3 N.Y.U. J.L. & LIBERTY 516, 517 (2008) ("The rule of law, liberalism, and democracy are often thought to make a happy triumvirate”). Tamanaha goes on to critique this traditional understanding of the relationship between democracy on the one hand and rule of law and liberalism on the other. Id. at passim. 330. See id. at 543-47. 331. SCHMITT, supra note 187. 154 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 of Russia to President Donald Trump have attacked rule-of-law liberalism in those terms.332 As Francis Fukuyama points out, this populo-nationalist attack is nothing new: German proto- nationalists like Paul de Lagarde accused liberalism (and its rule of law rationalism) as causing the decay of national community and national identity on similar grounds.333 The critique of presumptions thus radiates beyond “simple” rules of evidence. It affects if the rule of law is a competitor for how world society constructs its own reality or if it can be a means to mediate between competitors in a fair and even-handed manner.

A. The Theoretical Foundation of Presumptions A full understanding of the theoretical problem posed by presumptions requires an inquiry into their theoretical foundations. The doctrinal baseline for presumptions in all forms of international justice is the principle of good faith.334 The baseline assumption was first articulated in international law.335 It has since been adopted in disputes between states and non-state actors and in transnational commercial disputes.336 International courts and tribunals look to the principle of good faith as a default rule of decision.337 All else being equal, the parties are assumed to have acted in good faith.338 In other words, it must be proven that the parties acted with less than good faith.339 This rule supplements the use of burdens of proof: a party can sustain its burden of proof even in the absence of positive evidence as to what occurred.340 All it must do is show that the allegation by the party opponent would assume bad faith on the part of the moving party.341 In the absence of factual evidence either way, the

332. FRANCIS FUKUYAMA, IDENTITY: THE DEMAND FOR DIGNITY AND THE POLITICS OF RESENTMENT 9, 158-9 (2018). 333. Id. at 65-66, 70. 334. SOURGENS ET AL., supra note 87, at 117. 335. See CHENG, supra note 12, at 106 (“Contracting parties are always assumed to be acting honestly and in good faith.”). 336. SOURGENS ET AL., supra note 87, at 117; TRANS-LEX.ORG, supra note 271. 337. See CHENG, supra note 12, at 106. 338. Id. 339. Id. 340. Id. 341. See id. 2020] TRUTHS IN TRANSLATION 155 moving party can then rest on the assumption that it would have acted in good faith.342 The Corfu Channel case discussed throughout the Article is one example of this use of good faith presumptions in action. The International Court of Justice in Corfu Channel refused to conclude that Albania colluded with Yugoslavia to mine the Corfu straights in the absence of positive proof to that effect.343 It did so because such collusion would be bad faith—it would be tantamount to an admission of a violation of the law of innocent passage.344 At the same time, the Court ruled that Albania would be well aware of the laying of mines by third parties in its strategic waterways and therefore would have been able to warn third parties about the presence of the mines.345 It imputed actual knowledge of the mines from the general practice of states to observe their strategic waterways together with past evidence that Albania had acted consistently with this practice on previous occasions.346 To ground presumptions in good faith means that presumptions can be divided into the two familiar elements of good faith.347 The first element of good faith is parties act with honesty in fact.348 Consequently, when good faith becomes the basis for a presumption, it leads the decisionmaker to assume that the party invoking the presumption acted with honesty in fact.349 This means that the decisionmaker will assume that motives of that party are pure of heart rather than malicious.350 And the decisionmaker will also assume that the contemporaneous representations made by the party are truthful rather than fraudulent.351

342. See id. 343. Corfu Channel (U.K. v. Alb.), Judgment, 1949 I.C.J. Rep. 4, 16-18 (Apr. 9). 344. Id. 345. Id. at 18-20. 346. Id. 347. SOURGENS ET AL., supra note 87, at 117. 348. Id. 349. See CHENG, supra note 12, at 106; HICEE B.V. v. Slovk., PCA Case Repository No. 2009-11, Partial Award, ¶ 129 (May 23, 2011); ADC Affiliate Ltd. v. Hung., ICSID Case No. ARB/03/16, Award, ¶ 475 (Oct. 2, 2006); Gemplus S.A. v. United Mexican States, ICSID Case No. ARB(AF)/04/2, Award, ¶ 4-142 (June 16, 2010). 350. See CHENG, supra note 12, at 106; HICEE B.V., PCA Case Repository No. 2009-11 at ¶ 129; ADC Affiliate Ltd, ICSID Case No. ARB/03/16 at ¶ 475. 351. Gemplus S.A., ICSID Case No. ARB(AF)/04/2 at ¶ 4-142. 156 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1

The second element of good faith is reasonableness.352 Consequently, when good faith becomes the basis for a presumption, it leads the decisionmaker to assume that the party invoking the presumption acted reasonably.353 This means that the decisionmaker will assume that the party invoking the presumption acted consistently with applicable custom or practice.354 In other words, it will substitute proof of the practice itself and proof of general adherence by the invoking party with the practice for proof that the practice was followed in the specific instance at bar. This use of good faith as the basis for presumptions creates a potential problem. It logically requires the decisionmaker to take sides before hearing the evidence. The actor’s good faith is presumed. This entails that the audience’s or “victim’s” perspective is discounted. There is no record basis for this choice. Yet, it is frequently dispositive of core disputed factual questions. In international law, it means that states are presumed to have acted in accordance with general expectations of states in similar circumstances.355 This risks discounting the perspective of states in similar position to the audience or victim states (consider a dispute between a nuclear weapon state such as North Korea and a non-nuclear weapon state such Japan regarding nuclear weapons tests). In transnational commercial law, companies are presumed to have acted in accordance with trade practice in their respective trade, accounting for size, place and industry.356 This presumption risks discounting the perspective of their trading partners if these partners are smaller, from a different place, or industry. Investor- state arbitration showcases this risk with even greater clarity as it risks discounting the perspective of the state regarding investor conduct and vice versa.357

352. SOURGENS ET AL., supra note 87, at 117. 353. Micula v. Rom., ICSID Case No. ARB/05/20, Award, ¶ 673 (Dec. 11, 2013); Phx. Action Ltd. v. Czech, ICSID Case No. ARB/06/5, Award, ¶ 127 (Apr. 15, 2009); CDC Grp. Ltd. v. Republic of Sey., ICSID Case No. ARB/02/14, Award, ¶ 47 (Dec. 17, 2003); BRIDAS SAPIC v. Turkm., Case No. 9058/FMS/KGA, Partial Award, ¶ 139, n.13 (ICC Int’l Ct. Arb. 1999). 354. See supra note 352. 355. CHENG, supra note 12, at 107. 356. Compare TRANS-LEX.ORG, supra note 271 (governing presumptions) with Principle I.1.1(b), TRANS-LEX.ORG, https://www.trans-lex.org/901000 [https://perma.cc/F9F6-LFUN] (last visited Sept. 10, 2020) (codifying context-sensitive good faith obligations). 357. SOURGENS ET AL., supra note 87, at 117-20. 2020] TRUTHS IN TRANSLATION 157

This theoretical basis of presumptions thus provides a more nuanced explanation why the Kim v. Uzbekistan tribunal chose its presumptions according to which it resolved the dispute as it did. Uzbekistan asked the Kim tribunal to find that the investor had acted in a corrupt manner.358 This conclusion concerned the actions of a commercial party.359 The presumption of good faith defaults to the relevant practice of the acting party—that is the commercial party. As the Kim claimants could show that there was an innocuous trade practice to structure investments in Uzbekistan in a manner similar to the way that the Kim claimants had done, the Kim claimants received the benefit of the presumption.360 The Kim claimants received the benefit of the presumption despite the fact that from the state’s perspective, the evidence was similarly consistent with the perception of corruption.361 Presumptions bring about this result because the state was not the actor and its perspective therefore is ultimately not material.362

B. Presumptions as Narrative It should be reasonably straight forward that this use of good faith in presumptions will lead to problems. It is not a compelled conclusion that good faith must favor the actor’s perspective. An analogy to the law of contracts readily shows why. A fictional law professor might tell her class “whoever of you dunces can answer my next question correctly gets an A in my class.” A fictional law student could rise to the challenge and correctly answer the question. Did the law professor make an offer or did she tell a joke? This question critically depends upon perspective. From the perspective of most (non-fictional) law professors, it is fair to assume that the statement is a joke. But when the author offered the statement as a hypothetical question after teaching Lucy v. Zehmer to introduce some levity into class discussion, all students generally considered the statement to be a serious offer (and for

358. Kim v. Republic of Uzb., ICSID Case No. ARB/13/6, Decision on Jurisdiction, ¶ 546 (Mar. 8, 2017). 359. Id. 360. Id. at ¶ 547. 361. Id. at ¶ 589. 362. SOURGENS ET AL., supra note 87, at 117-20. 158 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 that matter, not very funny).363 Speaker’s meaning and audience meaning diverge on the critical question of what happened. Lucy v. Zehmer classically sides with the ordinary meaning of the audience of a statement.364 The Restatement (Second) of Contract makes this link explicit when it deems a manifestation of willingness to be an offer when it is “so made as to justify another person in understanding that his assent to that bargain is invited and will conclude it.”365 The US common law of contracts as summarized in the Restatement would thus conclude that our fictitious law professor was serious and did not make a joke. Problematically, if one analyzes the same hypothetical through the lens of presumptions at work in international dispute resolution one arrives at the opposite result. There is no additional record evidence to determine whether the fictitious law professor was joking. This means that the fictitious law professor could invoke a presumption. That presumption would look to other similarly situated actors as a frame of reference.366 Our example posits that law professors, unlike law students, would have understood the statement as a joke. Because the focus of the presumption analysis is on the actor, the statement would therefore be treated as a joke. Tellingly, both regimes seek to reduce what happened to a similar frame of refence: reasonableness. The objective theory of mutual assent informing the US common law of contracts looks to establish what a reasonable person in the position of the offeree would have understood.367 The force of the objective theory is factual—it assumes that it is more likely to arrive at the true ex ante expectations of the parties.368 This understanding of reasonableness and correctness is frequently coupled with a latent reference to good faith.369 The theory underlying presumptions

363. See Lucy v. Zehmer, 196 Va. 493 (1954). 364. Id. at 502. 365. RESTATEMENT (SECOND) OF CONTRACTS § 24 (AM. LAW INST. 1981) (emphasis added). 366. See supra note 353. 367. See Lawrence M. Solan, Contract as Agreement, 83 NOTRE DAME L. REV. 353, 401 (2007) (discussing Justice Cardozo’s contract jurisprudence and “the ex ante intent of the contracting parties”). 368. Id. 369. See id. (noting the link between the ordinary meaning and good faith jurisprudence). 2020] TRUTHS IN TRANSLATION 159 similarly looks to reasonableness as its benchmark.370 Presumptions are clearly a result of the principle of good faith,371 and assume that it is more likely than the alternative to arrive at the true contemporaneous understanding of the events at bar.372 This means that the difference in result cannot be explained away simply by the different legal contexts in which the different results arise. Both legal contexts intend to do the same thing—reach the reasonable result—but then reach the exact opposite result. How is this possible? The use of good faith and reasonableness in both contexts ultimately reduces the question of what happened to one of narrative. What happened is not a question of establishing something about the “real world.” In both instances, the ultimate determination of whether the law professor joked or not is “constructed” to justify and compel a certain result (in this case, the result would be whether the student answering the question should receive an A).373 We achieve this result by super-imposing discursive conventions over the world as we cannot learn more about the world itself.374 This means we are ultimately asking a different question. We did not seek to establish whether the law professor actually joked. Nor did we seek to establish if the student answering the law professor’s question actually thought he or she would receive an A in the class. Instead we seek to superimpose a uniform objective matrix over the evidence to draw a conclusion as to how a legal dispute should be decided.375 This objective matrix can be threateningly circular. The presumption that the parties acted in good faith overlaps with the obligation that the parties ought to treat each other with good faith. International law imposes an obligation upon states to act honestly in fact and to act reasonably.376 Transnational commercial law imposes much the same obligation for commercial actors.377 In both cases reasonableness is construed against the same

370. See infra note 353. 371. Id. 372. SOURGENS ET AL., supra note 87, at 117-20. 373. JACQUES DERRIDA, DE LA GRAMMATOLOGIE 54-55 (1967). 374. Id. 375. JEAN BAUDRILLARD, SIMULCRA AND SIMULATIONS 7 (1994). 376. CHENG, supra note 12, at 105-62. 377. Principle I.1.1, TRANS-LEX.ORG, https://www.trans-lex.org/901000 [https://perma.cc/F9F6-LFUN] (last visited Sept. 10, 2020). 160 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 background condition of uniformity with general practice of like situated actors that is at work in the use of presumptions.378 What is and what ought to be thus overlap. Factual findings premised in a presumption of good faith construct events out of evaluative claims. Good faith is a normative concept rather than descriptive one. To state that a party acted in good faith is to evaluate conduct. To state that a party acted in bad faith is to blame that party for wrongdoing. To assume that a party acted in good faith thus constructs facts from a source external to the events. This use of presumptions again runs into a version of the is- ought problem.379 The is-ought problem submits that there is a difference in kind between descriptive statements and evaluative statements.380 Evaluative statements are premised upon norms.381 These norms cannot be derived from facts alone.382 They require an ulterior source standing behind facts.383 For example, one may observe that Romans eat fish on Friday. This would be a descriptive statement. This statement becomes evaluative or normative when coupled with the command “when in Rome, do as the Romans.” This norm—act uniformly—is not a factual description. It requires external validation. By way of analogy, it is again instructive to look to US contract law. Early efforts by Karl Llewellyn to codify the US commercial law in the Uniform Commercial Code faced headwind with regard to the definition of the obligation to treat contractual counterparties with good faith.384 Karl Llewellyn sought to introduce an obligation into the Uniform Commercial Code that merchants treat each other according to the reasonable commercial standards of fair dealing in the trade.385 This part of

378. CHENG, supra note 12, at 134-36; TRANS-LEX.ORG, supra note 377. 379. DAVID HUME, A TREATISE OF HUMAN NATURE 469-70 (P.H. Nidditch ed., 2d ed. 1978). 380. Id. 381. See Rachel Cohon, Hume’s Moral Philosophy, STAN. ENCYCLOPEDIA MORAL PHIL. (Aug. 20, 2018), https://plato.stanford.edu/entries/hume-moral/#io [https://perma.cc/YDC3-DFDC] (outlining the different interpretations of the problem in light of evaluative/ normative and descriptive statements). 382. HUME, supra note 379, at 455-70. 383. Id. 384. Peter Winship, Jurisprudence and the Uniform Commercial Code: A Commote, 31 SW. L.J. 843, 856 (1977). 385. Id. at 855. 2020] TRUTHS IN TRANSLATION 161 the implied covenant of good faith codified the reasonableness requirement is akin to the statement “when in Rome, do as the Romans.”386 Llewellyn’s efforts failed.387 The American Bar Association’s Section on Corporation, Banking and Business law resisted inclusion of the reasonableness requirement insisting that there only be a good faith obligation to act with honesty in fact.388 This episode highlights that uniformity of conduct does not create an obligation of its own accord. If that had been the case, the objection of the American Bar Association would have been defeated as absurd when it was made. Instead, it took until the 2001 revisions to the Uniform Commercial Code that reasonable standards in the trade in fact create an obligation of fair dealing as part of the general definition of good faith in Article 1 of the Code.389 Presumptions run into the is-ought problem in reverse. Presumptions are grounded in the principle of good faith.390 The principle of good faith is a legal command as to how an actor ought to behave.391 Presumptions deploy this principle to establish how the actor did behave.392 Presumptions infer an “is” from an “ought.” Transnational commercial practice shows why this could well be a problem. Good faith in global commerce relies heavily upon self- regulation.393 International merchants and their lawyers establish between themselves what best practices they ought to follow when doing international business.394 These best practices then mature through adoption into a kind of running code for the global business community.395 One need not be too hard-nosed of a realist to understand that such transnational commercial practices are frequently

386. Id. 387. Allen R. Kamp, Uptown Act: A History of the Uniform Commercial Code: 1940-49, 51 SMU L. REV. 275, 280 (1998). 388. Clayton P. Gillette, Limitations on the Obligation of Good Faith, 1981 DUKE L.J. 619, 623 (1981). 389. Margaret L. Moses, The New Definition of Good Faith in Revised Article 1, 35 UCC L.J. 47, 47 (2002). 390. SOURGENS ET AL., supra note 87, at 117-20. 391. CHENG, supra note 12, at 134-36; TRANS-LEX.ORG, supra note 377. 392. SOURGENS ET AL., supra note 87, at 117-20. 393. KLAUS PETER BERGER, THE CREEPING CODIFICATION OF THE NEW LEX MERCATORIA 44- 46, (2d ed. 2010). 394. Id. 395. Id. 162 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 aspirational. They are expressions of intent. As such they are meaningful normative yardsticks. They express a desire and consent to act according to such a set of agreed upon norms.396 However, this desire or consent does not always overlap with realities on the ground: the best practices harden into norms because parties include best practices into contracts with fellow merchants as obligations.397 Needless to say, such an inclusion would not be needed if the parties already could be depended upon to act consistently with the best practices in the ordinary course; by way of example, form contracts do not require business people to own a computer even when notices can be given by email. Such rudimentary facts of life—important as they may be—are assumed. Compliance with best practice is another matter.398 This makes commercial good faith obligations a dangerous tool to establish what a particular commercial party did on a particular occasion. Best practices, uncharitably, may very well resemble New Year’s resolutions. They express a sincere desire to act. Yet, to find out whether a person ran on July 4 of a given year, it would be dangerous to look to their new year’s resolution to run every day. We can now see the “reverse is-ought problem” underlying the application of presumptions. They project the self-regulatory intentions of states and commercial actors to determine that they in fact acted consistently with their aspirations. They thus base factual findings in a narrative of voluntary compliance with self- regulatory pronouncements. This premise is a “narrative” in the sense that it creates an arc from best intention to best practice to actual practice. It tells the story we wished were true about us. Needless to say, the underlying normative aspirations rarely are fully realized.

C. The Bias Problem—How to Test Narratives? The currently prevalent use of presumptions has a bias problem. As discussed in the last section, presumptions are premised in narratives. These narratives presuppose that actions in particular instances comport with normatively grounded expectations of how actors ought to act. This use of narratives is

396. Id. 397. Id. at 44. 398. Id. 2020] TRUTHS IN TRANSLATION 163 not premised in a descriptive frame of reference but instead imports a normative frame of reference to establish what actually happened in a given case. The choice of perspective from which to construct the presumptions now becomes acutely problematic. It determines whose narrative governs the fact-finding process. As we have seen in the context of the Kim v. Uzbekistan example, the choice of perspective focuses on the actor to whom the presumption would be applied rather than on its counter-party.399 It creates presumptions premised upon how the actor’s own immediate environment reconstructs its own aspirations and translates these aspirations into cognizable best-practice efforts.400 It therefore anchors what happened in the frame of reference of the very person whose conduct is at issue. This choice of perspective uncritically adopts the biases of the party whose conduct is at issue. Self-regulatory processes appropriately begin from the shared cognitive baselines between actors participating in them.401 These shared cognitive baselines rationalize a set of behavior in accordance with the evaluative aspirations of the group.402 These aspirations are an appropriate yardstick between the participants in the process. As participants in the process, they share the same authoritative expectations. The problem is that these self-regulatory expectations are used as a means to construct facts in disputes with outsiders. This is most clearly the case in proceedings like the Kim arbitration.403 These proceedings involve states and commercial parties. However, they rely upon the in-group’s assessment of its own conduct as a basis for factual determination despite the fact that the out-group has not participated in its self-regulatory processes and does not share the underlying authoritative expectations giving rise to them. Commercial parties have a very different view whether government agencies act capriciously than the government agencies themselves. Moreover, government agencies

399. Kim v. Republic of Uzb., ICSID Case No. ARB/13/6, Decision on Jurisdiction, ¶ 546 (Mar. 8, 2017). 400. BERGER, supra note 393, at 44-46, 285-92. 401. Id. 402. See GRALF-PETER CALLIESS & PEER ZUMBANSEN, ROUGH CONSENSUS RUNNING CODE A THEORY OF TRANSNATIONAL PRIVATE LAW 125 (2012). 403. See Kim, ICSID Case No. ARB/13/6 at ¶ 546. 164 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 may have similar misgivings about “corporate greed.” Yet, these differences in perspectives are silently glossed over. The problem is not limited to such hybrid proceedings. They can similarly arise in the context of state-to-state proceedings. To begin with, these proceedings themselves may take on a hybrid character such as when states espouse the claims of their nationals.404 Yet, even in classic state-to-state claims, narratives too frequently build up. Events such as the have demonstrated how such narratives almost led to nuclear war.405 Events like the aftermath of the assassination of Arch-Duke Franz-Ferdinand in Sarajevo in 1914 further demonstrate how narratives can lead Great Powers to enter into a world war.406 The use of presumptions thus risks organizing materials according to a matrix of alternative facts.407 Alternative facts at heart are descriptive claims that are premised in narrative-driven normative demands.408 The claim by President Trump regarding the size of the crowd witnessing his inauguration is such an alternative fact.409 It is premised in the normative demand of legitimacy.410 The narrative of legitimacy demands supporters. This narrative thus constructs the descriptive claim of large inauguration crowds. More to the point, the “Rules for Radicals” approach to politicized discourse requires that discourse be placed in the context of the experience of one’s own people to the exclusion of the opponent.411 Rules for Radicals type-manuals thus uses an approach to narrative-based discourse that refuses to yield the

404. See Ahmadou Sadio Diallo (Republic of Guinea v. Dem. Rep. Congo), 2010 I.C.J. Rep. 639, ¶ 58 (Nov. 30). 405. THE KENNEDY TAPES: INSIDE THE WHITE HOUSE DURING THE CUBAN MISSILE CRISIS 387-88 (Ernest May & Philip Zelikow eds., 2002). 406. CHRISTOPHER CLARK, SLEEPWALKER: HOW EUROPE WENT TO WAR IN 1914 560-62 (2012). 407. Strong, supra note 2, at 137-39; Julie Beck, This Article Won’t Change Your Mind, The Facts on Why Facts Alone Can’t Fight False Beliefs, ATLANTIC (Mar. 13, 2017), https://www.theatlantic.com/science/archive/2017/03/this-article-wont-change-your- mind/519093/ [https://perma.cc/M8M9-K4JQ]. 408. See Strong, supra note 2 at 137-39. See also Beck, supra note 407. 409. See Jim Rutenberg, ‘Alternative Facts’ and the Costs of Trump-Branded Reality, N.Y. TIMES (Jan. 22, 2017), https://www.nytimes.com/2017/01/22/business/media/alternative-facts-trump- brand.html [https://perma.cc/U5W9-RY9G]. 410. Strong, supra note 2, at 137-39. 411. ALINSKY, supra note 128, at 127. 2020] TRUTHS IN TRANSLATION 165 frame for appraisal and construction of what counts as “facts.” It is this radicalization which ultimately underlies the us/them dynamic of the post-truth syndrome—“we” becomes the arbiter of truth by anchoring valid statements in a shared—and intentionally weaponized—sense of community and communal narrative.412 The problem of presumptions is that they risk operating in the same manner. If one is not careful, the use of presumptions anchors the construction of facts in the narratives of the very people pacific dispute resolutions needs to police. These narratives similarly are premised in normative demands. Thus, states frequently make the normative demand that they do not torture dissidents.413 International dispute resolution processes, and by extension the international rule of law which relies upon them, frequently construct facts on the basis of descriptive claims consistent with this demand by sweeping such allegations under the rug or treating them as uncorroborated.414 Presumptions thus are powerful tools to resolve disputes—but they are a tool that can come at cost: the specter of alternative facts.

V. PRESUMPTIVE POLYVALENCE

A. Presumptions as Translation In order to overcome the problem of presumptions, we must begin with a functional analysis. How would the rule of law meet its function to transcend community-based narrative given the reality of incomplete information? This reality forecloses any call for the abolition of presumptions. It also should caution against

412. FUKUYAMA, supra note 332, at 88. 413. Kazakhstan Has ‘Zero Tolerance’ Policy on Torture, Officials Tell European Parliamentarians, ASTANA TIMES (Apr. 17, 2018), https://astanatimes.com/2018/04/kazakhstan-has-zero-tolerance-policy-on-torture- officials-tell-european-parliamentarians/ [https://perma.cc/BUK2-3D9T]; In Dialogue with Uzbekistan, Committee against Torture Experts Urge Effective Implementation of Legislation and Further Investigations into Rights Violations, OHCHR (Nov. 13, 2019), https://www.ohchr.org/en/NewsEvents/Pages/DisplayNews.aspx?NewsID=25294&Lan gID=E [https://perma.cc/7FAA-AC4H]; Stephanie Nebehay, U.N. launches investigations into killings, torture in Venezuela, REUTERS (Sept. 27, 2019), https://www.reuters.com/article/us-venezuela-security-rights/u-n-launches- investigations-into-killings-torture-in-venezuela-idUSKBN1WC12H [https://perma.cc/GDK4-GB72]. 414. Caratube Int’l Oil Co. v. Kaz., ICSID Case No. ARB/08/12, award (May 30, 2012), ¶¶ 172-5. 166 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 disfavoring presumptions. The reality is that global dispute resolution bodies of any kind must act without a record that would satisfy later generations of historians looking into the same question. Abolishing presumptions would return one to the technical applications of burdens and standards of proof. The mechanical use of burdens and standards of proof, as discussed already, is a deeply unsatisfying solution.415 The resolution of disputes would then depend purely upon the procedural posture of the case.416 This in turn would do little to raise confidence in the ability of global dispute resolution processes fairly to resolve disputes. Or differently put, a resolution premised in alternative facts is preferable to a recognition of decisional impotence. In fact, it was such decisional impotence which undergirded the “war paradigm” of the colonialist period in international law: as all else fails, might makes right.417 If might makes right, the instinct of National- Socialist legal theorist Carl Schmitt becomes a self-fulling prophecy: political struggle becomes mortal struggle.418 What is more, presumptions also are premised in an appreciation of facts. One must prove that a presumption is applicable. This means one must prove a predicate for the presumption.419 Think again of proof of habit to stand in for proof of conduct on a specific occasion.420 To benefit from an inference from habit, one must first prove the habit in question. This substitutes proof of one fact (conduct on a specific occasion) with proof of another (habit).421 This means that presumptions are factually contestable.422 It is always possible to poke holes in the proof of a habit. Likewise, it is possible to prove that the habit does not cover what fact the party means to prove by it. For example, a habit to put on a seat belt does not prove that a person indicated their turn. Moreover, it

415. See Sections II.A.3 & II.B., supra. 416. See HERSCH LAUTERPACHT, THE FUNCTION OF LAW IN THE INTERNATIONAL COMMUNITY 105 (rev. ed. 2011) (making a similar observation in a different context). 417. See HATHAWAY & SHAPIRO, supra note 13, at 80-81. 418. SCHMITT, supra note 187, at 10. 419. See Section III.C. 420. Id. 421. Id. 422. PHILIP PETTIT, REPUBLICANISM: A THEORY OF FREEDOM AND GOVERNMENT 185 (1997). 2020] TRUTHS IN TRANSLATION 167 is possible to appreciate the habit in its richer factual context. This means that presumptions have an important part to play to return disputes about value to factually cognizable disagreements in the first place. At the same time, the current problem with presumptions is that they do not always resolve disputes. Rather, as discussed in the previous section, presumptions can import external narratives into the dispute resolution process. These narratives can in turn amplify the underlying dispute rather than resolving it—i.e., they can shield the perpetrator of a wrong from liability by invoking its narrative as a reason for decision as opposed to checking the narrative against fact. This kind of narrative based decision- making is similarly problematic. Returning to somewhat tautological first principles, the point of global dispute resolution processes is to resolve disputes. In order to resolve disputes, global dispute resolution processes must rely upon presumptions in order to address the endemic information deficit problem. At the same time, global dispute resolution processes cannot resolve a dispute by merely amplifying rather than testing the normative claims inherent in the use of presumptions. Dispute resolution ultimately requires an explanation by an international court or tribunal why the losing party failed to prevail.423 In the first place, this means that the losing party must understand the reasoning of the international court or tribunal.424 Doctrinally, this merely requires that the parties can understand the logic of the decision—whether or not they agree with its basis.425 To resolve the dispute, as opposed to determining a winner in a litigious contest, the decision must do more. Dispute resolution, taken literally, is to find a new solution for the dispute that is to re- solve its underlying problem. To achieve this, the losing party must accept the result as legitimate.426 At a minimum, the dispute

423. R. DOAK BISHOP & SILVIA MARCHILI, ANNULMENT UNDER THE ICSID CONVENTION 153- 54 (2012). 424. Mar. Int’l Nominees Establishment v. Guinea, ICSID Case No. ARB/84/4, Decision on Annulment, ¶¶ 5.08-5.09 (Dec. 22, 1989). 425. Id. 426. See, e.g., CHENG, supra note 12, at 291; Casado v. Chile, ICSID Case No. ARB/98/2, Annulment (Dec. 18, 2012), ¶ 266; Wena Hotels Ltd. v. Egypt, ICSID Case No. ARB/98/4, 168 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 resolution literature would demand respect for the autonomy of the parties.427 Respecting autonomy in turn demands that the parties be consulted in the decision-making process.428 This consultation must be visible on the face of the decision itself.429 The decisionmaker must provide an explanation that places the factual claims of the prevailing party into the context of the losing party so that the losing party can see that it was, in fact, heard rather than merely given an opportunity to speak.430 It must explain the decision in terms that the losing party would find authoritative.431 In theoretical terms, the international court or tribunal must provide such an explanation without offending the dignity interests of the losing party. Fukuyama recently theorized this dignity interest in the context of identity politics driving narrative- based alternative-fact-focused political decision-making.432 In this context, he theorized that the dignity could be split into two—a demand for equal respect, which he linked to the Greek notion of isothymia, and a demand for greater recognition premised upon perceived merit, which he linked to the Greek notion of megalothymia.433 Fukuyama submits that a key driver of current identity-based politics is a claim by a growing number of groups that the current paradigm of world order fails to satisfy their demand for isothymia, thus creating an identity-based merits claim that in turn falls into megalothymia.434 This disregard for equal dignity becomes a vicious spiral by feeding demands premised in megalothymia that in turn could destroy deliberative mechanisms premised upon the equal dignity of discourse participants.435

Annulment (Jan. 28, 2002), ¶ 57; Fraport AG v. Phil., ICSID Case No. ARB/03/25, Annulment (Dec. 23, 2010), ¶ 202. 427. ROGER FISHER & DANIEL SHAPIRO, BEYOND REASON 72-93 (2005). 428. Id. at 84. 429. See id. at 82-84. 430. See, e.g., CHENG, supra note 12, at 291; Casado v. Chile, ICSID Case No. ARB/98/2, Annulment (Dec. 18, 2012), ¶ 266; Wena Hotels Ltd. v. Egypt, ICSID Case No. ARB/98/4, Annulment (Jan. 28, 2002), ¶ 57; Fraport AG v. Phil., ICSID Case No. ARB/03/25, Annulment (Dec. 23, 2010), ¶ 202. 431. W. MICHAEL REISMAN, THE QUEST FOR WORLD ORDER AND HUMAN DIGNITY IN THE TWENTY-FIRST CENTURY: CONSTITUTIVE PROCESS AND INDIVIDUAL COMMITMENT 81 (2012). 432. FUKUYAMA, supra note 332, 9-10, 21-22, 95, 140-144. 433. Id. at 21-22, 95. 434. Id. at 85, 95, 140-144. 435. Id. 2020] TRUTHS IN TRANSLATION 169

Isothymia has an immediate application in dispute resolution. To reframe the discussion above slightly, the losing party must feel that it was treated as an equal of its opponent in the resolution of the dispute.436 A decision that offends the basic requirements of equal treatment would denigrate the losing party by implying that it is inherently not worthy of basic equal in treatment.437 Such treatment would offend basic dignity interests and as such be potentially illegitimate.438 International dispute resolution is conscious of this dignity interest to isothymia. It requires that a court or tribunal treat the parties with formal equality and substantive equality of arms.439 That means that the court’s or tribunal’s explanation why a party failed to prevail cannot be one that ties a loss to status alone (e.g., one party is not a state and states are always right). 440 It must provide non-status-based reasoning to support a factual conclusion.441 Problematically, while conscious of this dignity interest, the remedies available to protect it in dispute resolution are highly limited, requiring close to a wanton disregard for the equality of the parties by a tribunal.442 In the context of findings of fact, these interests are notoriously under-enforced in the interest of protecting the discretion of the finder of fact to weigh the evidence.443

436. See, e.g., CHENG, supra note 12, at 291; Casado v. Chile, ICSID Case No. ARB/98/2, Annulment (Dec. 18, 2012), ¶ 266; Wena Hotels Ltd. v. Egypt, ICSID Case No. ARB/98/4, Annulment (Jan. 28, 2002), ¶ 57; Fraport AG v. Phil., ICSID Case No. ARB/03/25, Annulment (Dec. 23, 2010), ¶ 202. 437. See Thomas Wälde, “Equality of Arms” in Investment Arbitration, Procedural Challenges, in ARBITRATION UNDER INTERNATIONAL INVESTMENT AGREEMENTS – A GUIDE TO KEY ISSUES 161, 181 (Katia Yannaca-Small ed., 2010). 438. FUKUYAMA, supra note 332, at 9-10, 21-22. 439. MINE v. Guinea, ICSID Case No. ARB/84/4, Annulment (Dec. 22, 1989), ¶ 5.06; CHENG, supra note 12, at 291; Casado v. Chile, ICSID Case No. ARB/98/2, at ¶ 266; Wena Hotels Ltd. v. Egypt, ICSID Case No. ARB/98/4, at ¶ 57; Fraport AG v. Phil., ICSID Case No. ARB/03/25, at ¶ 202. 440. See Wälde, supra note 437, at 181. 441. MINE v. Guinea, ICSID Case No. ARB/84/4, Annulment (Dec. 22, 1989), ¶ 5.06; CHENG, supra note 12, at 291; Casado v. Chile, ICSID Case No. ARB/98/2, at ¶ 266; Wena Hotels Ltd. v. Egypt, ICSID Case No. ARB/98/4, at ¶ 57; Fraport AG v. Phil., ICSID Case No. ARB/03/25, at ¶ 202. 442. BISHOP & MARCHILI, supra note 423, at 132-35. 443. Alapli Elektrik BV v. Turk., ICSID Case No. ARB/08/13, Annulment (July 10, 2014), ¶ 234; Tulip Real Estate Investment & Development BV v. Turk., ICSID Case No. ARB/11/28, Annulment (Dec. 30, 2015), ¶ 150. 170 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1

Unfortunate, though, as this disconnect might appear at first sight, it helps to open the door for a legal understanding of the potential legitimacy deficit if presumptions are used “incorrectly.” The legal toolkit available to the international lawyer shows that as a legal matter, presumptions can be used incorrectly when a court or tribunal uses them in a manner that fails to meet the aspirations of isothymia. Yet, it also permits that this failure can frequently fall beyond review. It sets up precisely the space in which international legal decisions can be binding and beyond review on the one hand and potentially illegitimate on the other. How, then, does one use presumptions correctly? Thinking of presumptions in terms of isothymia here is helpful. Presumptions can make broader value claims standing behind them factually intelligible and contestable. Presumptions do so when their predicates are examined with care and inferences are narrowly drawn. Such use of presumptions means that the value claims standing behind presumptions are scrutinized in factual terms in their own right. Isothymia demands as much—it requires one to take seriously the value claims of those advancing a presumption and those resisting it. Therefore, to prefer a value claim without such scrutiny is to ignore the interests of the party against whom the presumption is invoked. Just as importantly, presumptions must be able to work and be invoked both ways. A decisionmaker must hear the presumptions that organize the same record evidence from all perspectives. That is, it must render a factual picture that is cognizant of the perspectives (literally) of both parties and make a careful determination of how these perspectives overlap and diverge–and how this overlap and divergence sheds light on the value claims they each advance. Isothymia is only heeded when the factual picture becomes complex; when we can see, in the words of historian Christopher Clark, how each party could come to see reality through its lens in a coherent manner and can engage with that perspective on its own terms.444 So used, presumptions become tools to translate between the various claims made by the parties. They can only do so by embedding the use of presumptions in a rich factual context. In this context, decisionmakers can understand how and why certain

444. CLARK, supra note 17, at 5. 2020] TRUTHS IN TRANSLATION 171 values are normatively salient. In this context too, however, decisionmakers can also understand how and why certain values shape the factual focus of our analysis and can expand their perspective to take a fuller view of the facts by taking into account the values and perspectives of the other disputing party.445 The analogy of translation is apt because there is no universal language through which the court or tribunal could run this translation.446 There is no universal narrative or precognitive bias shared irrespective of lived experience. Or differently put, the necessarily constructed nature of presumptions forecloses the use of the “real world” as a tiebreaker. The need for presumption arises precisely because there is insufficient information on what the real world looks like. All there are, are puzzle pieces and competing claims as to what the puzzle depicts. The task of presumptions in this sense resembles that of linguistic translation in a literal sense. Languages like few other human constructs are embedded with crisscrossing layers of cognitive narratives.447 Even between language communities with longstanding historical contacts, there are terms that are near untranslatable.448 English expressions such as “common sense” find no direct analogue in French, a language from which English borrowed heavily.449 “Bon sens”—a translation suggested for example in Collins Dictionary comes close.450 But “sense” is not in fact “common” but “good.” In fact, one French commentator theorizing on appropriately “common sense” in French notes that to accuse a person of such common sense (raison commune ou sens common) is to put in doubt their intellectual acumen.451

445. I build just such a presumption in a different article. See Frédéric G. Sourgens, The Precaution Presumption, 31 Eᴜʀ. J. Iɴᴛ’ʟ L. (forthcoming 2021). 446. GEORGE STEINER, AFTER BABEL 109-11 (3d ed. 1998). 447. Ludwig Wittgenstein, Philosophische Untersuchungen, in 1 WERKAUSGABE 225, 254 (Suhrkamp Verlag, 1984). 448. STEINER, supra note 446, at 382. 449. PHILIP DURKIN, A HISTORY OF LOANWORDS IN ENGLISH 171 (2014) (attesting to heavy borrowing in English from French). 450. Common Sense, COLLINS DICTIONARY, https://www.collinsdictionary.com/dictionary/english-french/common-sense [https://perma.cc/A8BN-66YR] (last visited Oct. 4, 2020). 451. Oscar Brenifier, Le bon sense est-il commun?, DIOTOME (2007), http://www.educ-revues.fr/DIOTIME/AffichageDocument.aspx?iddoc=32848 [https://perma.cc/ZTQ5-NZB2] (last visited Oct. 4, 2020). In a very French and somewhat 172 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1

“Common sense” in English has rather a less pejorative meaning: it refers to “sound and prudent judgment based on a simple perception of the situation or facts.”452 In reverse, “se débrouiller” is a French term for which there is no ready English translation.453 To get by or to muddle through might come close, but it does not capture the commonplace pride associated with the original word. Similarly, Shakespeare’s exuberance is nearly untranslatable into French despite the deep historical ties between both countries and languages.454 Further, the task of translation cannot be accomplished through recourse to a universal language.455 There is no common denominator through which translation could function. Linguistic translation, in other words, shares the problem we noted about presumptions that there is no “real language” or “real world” that could serve as a tiebreaker for meaningful expression.456 Particularly when dealing with highly complex, narrative-driven expression, translation has to accomplish its task through holistic comparison, case by case approximation, and experienced reinvention of the underlying material.457 Importantly for the current context, the seemingly insurmountable difficulties faced by translation have not stopped it in its tracks. Literary translation flourishes, and authors such as Shakespeare have deeply influenced foreign literary luminaries such as Goethe despite the apparent chasms between their linguistic home worlds.458 In turn, some of the best Nietzsche scholars today hail from the English-speaking world despite the fact that Nietzsche’s use of the German language is intimately

ironic sense, the author of the essay then proceeds to ground common sense by reference to academic philosophical reasoning. 452. Common Sense, MERRIAM-WEBSTER, https://www.merriam- webster.com/dictionary/common%20sense [https://perma.cc/64ZY-B9X2] (last visited Oct. 4, 2020). 453. French Word of the Day: se débrouiller, THELOCAL.FR (Dec. 14, 2018), https://www.thelocal.fr/20181214/french-word-of-the-day-se-dbrouiller [https://perma.cc/Q7LX-RWFL]. 454. STEINER, supra note 446, at 383-84. 455. Id. at 109-11. 456. Id. 457. See id. at 436. 458. NICHOLAS BOYLE, GOETHE, THE POET AND THE AGE 115-16 (1991). 2020] TRUTHS IN TRANSLATION 173 poetic.459 Fruitful communication, in other words, is possible despite the underlying differences between the linguistic experiences in different language communities. Rather than presenting obstacles to communication, this difference is an engine for creativity in both the source and the target language. There would be no Shakespeare without the Italian commedia del arte.460 There would be no German Urfaust without Shakespeare.461 The imported perspective rejuvenated the domestic idiom with new creative possibilities. The new perspective also allows different insights into the source texts that become visible only in translation. George Steiner’s masterful analysis of the history of the reception of Sophocles’ Antigone is one literary example of how translation discovers new meaning in a paradigmatic masterpiece of Western literature.462 The American translation of Italian humanist theory into constitutional action is meaningfully additive to any reading of Machiavelli’s Discourses on Livy.463 In a slightly less highbrow fashion, Matthew Pearl’s novel “The Dante Club” shows the influence of the translation of Dante’s inferno on the Holmes family (and on Oliver Wendell Holmes, later Mr. Justice Holmes of the US Supreme Court).464 Given this experience of translation, it is possible to surmise that a similar exercise might be possible in the task of conflict resolution. Negotiation scholars drawing on the work in peace talks suggest that a recontextualization of their respective problems in light of shared experiences is a successful means of communication between warring parties.465 The classical Getting to Yes describes one of the tools available to this end as

459. See generally KEITH ANSELL PEARSON, VIROID LIFE, PERSPECTIVES ON NIETZSCHE AND THE TRANSHUMAN CONDITION (1997); BRIAN LEITER, NIETZSCHE ON MORALITY (2002). 460. See Michele Marapodi, Introduction: Shakespeare’s Subversions, in SHAKESPEARE AND THE ITALIAN RENAISSANCE, APPROPRIATION, TRANSFORMATION, OPPOSITION 1-20 (Michele Marapodi ed., 2014) (discussing the influence of Italy on Shakespeare’s writing). 461. BOYLE, supra note 458, at 143-44, 219 (discussing the influence of Shakespeare on Götz von Berlichingen and the connection between Götz and Urfaust). 462. GEORGE STEINER, ANTIGONES, THE ANTIGONE MYTH IN WESTERN LITERATURE, ART, AND THOUGHT 202-03 (1984). 463. J.G.A. POCOCK, THE MACHIAVELLIAN MOMENT, FLORENTINE POLITICAL THOUGHT AND THE ATLANTIC REPUBLICAN TRADITION 509-52 (2d ed. 2003). 464. See generally MATTHEW PEARL, THE DANTE CLUB (2003). 465. ROGER FISHER & WILLIAM URY, GETTING TO YES, NEGOTIATING AGREEMENT WITHOUT GIVING IN 110-12 (3d ed. 2011). 174 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1

“negotiation jujitsu.”466 This jujitsu does not reject a position or argument, but recasts it in light of the “interests behind it [and . . . ] the principles it reflects.”467 This negotiation jujitsu allows a recontextualization of claims through translation by using the interests and principles as a reference point.468 Getting to Yes shows how this tactic worked in the context of the Egyptian-Israeli conflict.469 It showed how demands made by Egypt’s Nasser could be recast in a way that they would be unrealistic by reference to Nasser’s own principles.470 This jujitsu provides a tested means to translate that is not unlike linguistic contextual translation. This insight therefore should and can underpin a reconceptualization of presumptions to drive at the truth between the respective narratives of the disputing parties. If this project is successful, it again radiates beyond the narrow confines of the technical rules governing international dispute resolution. Rather, the rules governing factual finding in international dispute resolution would highlight the importance of the interstitial spaces between narratives making up the identity- based global policy discourses (be it theorized as a clash of civilizations or a return to identity politics).471 There is narrative and space between narratives. Dispute resolution would seek to claim this space between narratives as a “neutral” ground in which rule of law-based understandings of engagement can create shared realities. This ground would be neutral not in the sense that it is universal. Rather, it would be neutral in the sense that it is contextually shared. The point of this re-conception of presumptions is that we can accept that there is no objective truth to which all claims or narratives could be reduced and yet not give up the claim for neutrality inherent in the liberal rule of law. Much as there is no universal language, there is no universal reality. This does not stop us from translating between realities.

466. Id. 467. Id. at 111. 468. See id. 469. Id. at 112. 470. Id. 471. See generally SAMUEL P. HUNTINGTON, THE CLASH OF CIVILIZATIONS AND THE REMAKING OF WORLD ORDER (1996); FUKUYAMA, supra note 332. 2020] TRUTHS IN TRANSLATION 175

Or, from the point of view of the cultural historian, post-truth discourses amplify our sense of the peculiar—of the intensely cultural, or, as it has now become popular to say, intensely tribal construction of reality through distinction. This understanding of post-truth discourse merely reprises the juxtaposition of proto- romantic notions of people, or Volk, and its unique genius on the one hand and classical notions of rationality, or Verstand, and its universality on the other.472 This juxtaposition is not new—Kant and Herder died within a year of each other and represented much of the same dichotomy.473 The moniker of post-truth in this sense is merely the (pejorative) appraisal of romantic claims of identity viewed through the lens of enlightenment rationalism. Rule of law does not have to take sides in the opposition between peculiar and universal; in fact, and ironically, it becomes the more powerfully universal if it doesn’t. Specifically, the rule of law project can still thrive in this intensely peculiarized setting. Rather than focusing on a claim to shared rationalist universality, rule of law becomes a call to communication premised in translation. It would take at face value that the lived experience of each construction of reality is equally valuable and normatively valid. It would then seek to find contextually constructed shared meanings between these lived experiences. It would thus replace conceptions of international order premised in might or premised in universal truth or rationality with a conception of international order as premised in translation and communicative action. It is in this sense that rule of law as translation between realities would become more diffusely universal and more universally relevant than its rationalist counterpart: it is relevant whether or not one were to ascribe to a universal, objective (physical and moral) reality. This translative account of rule of law is particularly relevant in times in which the diffusion of cultural realities is acutely felt in political discourse. But it is no less relevant in times of relative calm as it would even in those times remind dominant paradigms to be mindful of the peculiar lived experiences of those to whom the rule of law is applied.

472. See STEINER, supra note 462, at 81 (discussing Herder and Kant). 473. See Michael Forster, Johann Gottfried von Herder, STAN. ENCYCLOPEDIA PHIL. (Aug. 25, 2017), https://plato.stanford.edu/entries/herder/ [https://perma.cc/6V9U-QJ6X]. 176 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1

B. Connotations and Presumptions The theoretical foundations of presumptions permit a conceptual recasting of presumptions that meets this functional demand. Presumptions are ultimately rooted in the principle of good faith. So far, the engagement with the principle of good faith has asked how a party acts in good faith. That is, it gave the classic doctrinal answer of good faith as honesty-in-fact and reasonable conduct. To achieve the necessary change in the conception of presumptions, one must switch from the doctrinal how to the jurisprudential why. What function does good faith serve? Why do parties have to act honestly and reasonably? The theoretical answer to this question is other regard.474 The point of good faith is to change purely self-regarding conduct towards cooperative conduct.475 Deals tend to work when the parties can trust each other’s faithful performance; they tend to fail in the face of constant self-dealing. Good faith mandates a minimum of faithfulness in performance to support and protect the viability of agreements. A switch from self-centered action to cooperative conduct requires a change in perspective. It requires one to regard or consider one’s counterparty, understanding its motivations and goals, and to act in a manner that is consistent with the newly formed joint enterprise. This understanding of good faith has deep doctrinal roots. For example, the law of treaties requires that parties perform treaties in good faith.476 As the doctrinal literature explains, this means that the text of a treaty cannot be read literally to achieve an unexpected advantage.477 Rather, the treaty must be read in accordance with its spirit or purpose.478 It thus must be read so as to achieve the goals of the common enterprise rather than to achieve self-gratification.479

474. See Frederic G. Sourgens, Reconstructing International Law as Common Law, 47 Geo. WASH. INT’L L. REV. 1, 52-55 (2015); see also David Pozen, Constitutional Bad Faith, 129 HARV. L. REV. 885, 955 (2016). 475. E. ALLAN FARNSWORTH, CONTRACTS 504-07 (3d ed. 1999). 476. Vienna Convention on the Law of Treaties art. 26, May 23, 1969, 1155 U.N.T.S. 331 (entered into force Jan. 27, 1980). 477. MARK VILLIGER, COMMENTARY ON THE 1969 VIENNA CONVENTION ON THE LAW OF TREATIES 367 (2009); RICHARD GARDINER, TREATY INTERPRETATION 151 (2008). 478. See supra note 476. 479. Id. 2020] TRUTHS IN TRANSLATION 177

International and transnational law also recognize this duty beyond the narrow confines of the performance of a treaty. Both international and transnational law prohibit parties from abusing their rights.480 The concept requires that when “the owner of a right enjoys certain discretionary power” the power “must be exercised reasonably, honestly, in conformity with the spirit of the law and with due regard to the interests of others.”481 One can theorize that a determination of whether a party acts in accordance with good faith requires a relative understanding of their respective expectations and interests created by the circumstances in which the parties find themselves as a whole.482 Good faith provides a means to translate the respective factual demands by the parties to cooperative benefits into legally cognizable claims. They do so in much the same way—if to a potentially different degree—as “negotiation jujitsu” discussed in the prior section: it asks the parties to apply the interests and principles underlying their own factual demand and apply these interests and principles to the position of counterparty to establish whether the original demand is ultimately inconsistent with the spirit of their agreement.483 If the interests and principles invoked by the original demand can be defeated by a stronger factual case when they are applied in reverse, the original demand is not made with honest conviction as it tends to deny the counterparty the cooperative benefit for which it bargained.484 The demanding party thus would have to modify its demand to account for the respective interests of its counterparty validated by the principles and interests underlying the demanding party’s own claim. This function-understanding of good faith allows for a reconceptualization of presumptions. As presumptions are premised in the principle of good faith, they must not just reflect its specific doctrinal elements of honesty-in-fact and reasonableness. Rather, they must fulfil the aspirations of good

480. Aerial Incident of 10 August 1999 (Pak. v. India), Judgment, 2000 I.C.J. 12, 30 (June 21); Phoenix Action Ltd. v. Czech, ICSID Case No. ARB/06/5, Award (Apr. 9, 2009), ¶¶ 93-95; Principle No. I.I.4, TRANS-LEX, https://www.trans-lex.org/906500 [https://perma.cc/X48Y-BUQK] (last visited on Sept. 7, 2020). 481. CHENG, supra note 12, at 133-34. 482. For a full discussion of this understanding of good faith, see Sourgens, supra note 474, at 52-55. 483. FISHER & URY, supra note 465, at 110-12. 484. Id. 178 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 faith as a translative tool to permit legally enforceable communication between differently positioned actors. The functional analysis of good faith itself permits a broader diagnosis. Any dogmatic deployment of presumptions in violation of principles of other regard and context-sensitive inquiry into the relative points of view of the parties is inimical to the principle of good faith. Such dogmatic use of presumptions would depend upon self-regard, not other-regard. This is precisely what the doctrine of good faith in international law and transnational law sought to avoid. Rather than forcing parties to engage with their counterparty and take the counterparties’ interests into account as required by the principle of good faith, the law of presumptions unwittingly picks the part of the doctrine of good faith that, in this new context, has the opposite effect. To say that the law of presumptions is premised in the principle of good faith therefore must reject this type of invocation as misguided. This means that it is necessary to re-think how presumptions may be used in light of their source principle of good faith. The point of good faith principles is that they require a matrix of normative translation. Honesty in fact is a requirement to avoid subterfuge and thus present a counterparty with a clear starting position.485 Reasonableness, in turn, requires both parties to communicate in a manner that is consistent with their community expectations while being mindful of the reasonable reliance interests of their counterparty.486 Good faith thus has multiple access points that each can lead to valid arguments. It allows for multiple ways in which it engage in discourses that are premised in incommensurable narratives. A way to express this point is to say that good faith, and thus presumptions, have to make sense across different connotations.487 The current understanding of good faith in presumptions was doctrinally denotating and did not allow for this linguistic nuance. It focused upon the narrative of the actor. Good faith as a translative tool must focus on the narrative of the actor, the counterparty, and those affected by the decision.488 Thus, good faith is no longer fixed in some objective denotation—nor depends,

485. FARNSWORTH, supra note 475, at 504-07. 486. Id. 487. See Sourgens, supra note 474, at 36-37. 488. Id. 2020] TRUTHS IN TRANSLATION 179 for the validity of its argument, on a single discourse or narrative. It must be cognizant of connotations or open to acting in multiple narratives at the same time.

C. Deploying Translating Presumptions The theoretical reconfiguration of presumptions as truly translative tools premised in the principle of good faith is only more than a fancy string of paronyms if it can actually be deployed. It is therefore important to provide a blueprint for the actual use of translating presumptions. This blueprint in many instances will lead to results that approximate current practice—that is they will not alter the result reached by the International Court of Justice in Corfu Channel or the investor-state tribunal in Kim v. Uzbekistan. Yet, in some instances, it will indeed change the fact-finding process in outcome-determinative ways. The starting point for the adaptation of translating presumptions is to use the toolkit that is already on display in the practice of presumptions. Thus, the practice of presumptions always contrasts the assumptions held by the actor (the Albanian state in Corfu Channel, the corporate investor in Kim) with the assumptions held by the counterparty (the British navy in Corfu Channel, the Uzbek anti-corruption officials in Kim).489 The practice of presumptions, therefore, is already mindful of the duality of perspectives at play in fact finding. The problem so far is that this duality of perspective leads us straight back to the narrative problem discussed in the previous section; decision is premised in the assumptions and aspirations of one party to the exclusion of the other. 490 Such a decision lacks legitimacy in the eye of the losing party because its choice of premise will simply be unacceptable to the losing party in most instances. Additional perspectives are needed to overcome this problem. When good faith encounters this problem in other contexts, it typically looks to a reasonableness of reliance interests as its

489. Corfu Channel, Judgment, 1949 I.C.J. 4, 16-18 (Apr. 9); Kim v. Republic of Uzb., ICSID Case No. ARB/13/6, Decision on Jurisdiction, ¶¶ 546-87 (Mar. 8, 2017). 490. See supra Part IV.B. 180 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 guide.491 Could one party understand that its own actions would have an effect on a party that approaches those actions from a different vantage point? Classically, this reliance understanding of good faith developed in international law in the context of the law of unilateral acts or unilateral declarations.492 Though its doctrinal underpinnings are technical, its overall gist is deeply intuitive in practice. Unilateral declarations create international legal obligations for the declarant to the extent that the declaration would reasonably lead the intended audience to believe that it may rely on the actor’s representations.493 The mechanism is one of reasonable reliance closely related to doctrines of estoppel. The classical case for the articulation of unilateral acts are the Nuclear Tests cases brought by Australia and New Zealand against France.494 The core issue in the Nuclear Tests was whether France was legally entitled to conduct atmospheric nuclear weapons tests in the South Pacific. The International Court of Justice did not reach that question.495 Rather, it focused on a different aspect of the case that the parties had not fully developed: the President of France had made public statements that France would halt atmospheric nuclear weapons tests.496 The International Court of Justice determined that this statement by the head of state created reasonable reliance interests in the world community that France would indeed halt its test program.497 In light of the circumstances of the case, the Court did not require proof of actual reliance so long as the reliance was in fact reasonable. 498

491. See Frederic G. Sourgens, Supernational Law, 50 VAND. J. TRANSNAT’L L. 155, 184- 95 (2017). 492. See Duncan B. Hollis & Joshua J. Newcomer, ‘‘Political” Commitments and the Constitution, 49 VA. J. INT’L L. 507, 525 (2009). 493. For a full discussion of unilateral acts, see Frédéric G. Sourgens, Keep the Faith: Investment Protection Following the Denunciation of International Investment Agreements, 11 SANTA CLARA J. INT'L L. 335, 383-87 (2013). 494. See generally Nuclear Tests (N.Z. v. Fr.), Judgment, 1974 I.C.J. 457 (Dec. 20). 495. Sir Arthur Watts, Nuclear Tests Cases, Max Planck Encyclopedia of Public International Law (Jan. 2007), https://opil.ouplaw.com/view/10.1093/law:epil/9780199231690/law- 9780199231690-e185 [https://perma.cc/P9B5-46RZ]. 496. Brigitte Bollecker-Stern, L’affaire des Essais nucléaires français devant la Cour internationale de justice, 20 Aɴɴᴜᴀɪʀᴇ Fʀᴀɴᴄ̧ᴀɪs ᴅᴇ Dʀᴏɪᴛ Iɴᴛ’ʟ 299, 330 (1974). 497. Id. 498. Id. 2020] TRUTHS IN TRANSLATION 181

To analogize the situation to domestic law, the law of unilateral acts thus resembles principles of promissory estoppel at work in the context of charitable subscriptions.499 Promissory estoppel canonically requires that both the actor knew that its actions would reasonably induce reliance and actual reliance occurred.500 In the context of a charitable subscription, promissory estoppel drops the requirement of actual reliance.501 The reason for this lesser requirement in the context of charitable subscriptions is in part grounded in the difficulty in proving actual reliance in the cases in point—that is, they are not the kind of actions that would readily bring about a change in action by the receiving party but have value to that party nonetheless.502 This rationale is reasonably similar to the unilateral act scenario: how would the world community rely to its detriment upon a representation that France would halt its atmospheric nuclear weapons testing? In particular, how would a non-nuclear weapons state rely on that statement? Reliance will be a very difficult matter to prove. This understanding of good faith can be applied to presumptions by adding two considerations. First, it must be known how each party perceived the conduct of the counter-party as part of its respective external narrative. Second, it must be understood whether the conduct gave rise to reasonable reliance interests. Should each actor itself have a reasonable understanding that its conduct would be understood by the counterparty in a certain way? Importantly, this understanding weaved its way into the analysis in Corfu Channel and in Kim v. Uzbekistan, if as an overdetermined reason for a presumption that already been established on other grounds. In Corfu Channel, the Court noted the engagement between the Albanian forces and the British navy when British vessels were in the straights in question on prior occasions.503 This engagement implicitly supports reasonable

499. RESTATEMENT (SECOND) OF CONTRACTS § 90(2) (AM. LAW INST. 1981). 500. Id. at§ 90(1). 501. Id. at §90(2). 502. See FARNSWORTH, supra note 475, at 91-101. 503. Corfu Channel, Judgment, 1949 I.C.J. 4, 19 (Apr. 9) (“This vigilance sometimes went so far as to involve the use of force : for esample the gunfire in the direction of the British Orion and Sztperb on May 15th 1, 946, and the shots fired at the U.N.R.R.A. 182 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 reliance interests: that is, it is reasonable for the British navy to believe that no military vessels would be in the straights unobserved. This is precisely the presumption from which the UK benefited in the case—but justified by reference not to the UK’s narrative alone.504 It can now be translated in Albania’s narrative by contextualizing the UK narrative in the context of Albania’s own actions and rationale. The Kim v. Uzbekistan tribunal similarly spent significant time to anchor its decision in a detailed exposition of Uzbek law.505 The presumption in favor of Kim was not only appropriate as a matter of international practice.506 It was in fact appropriate because the Uzbek law was itself not satisfied by the arguments raised by Uzbekistan.507 Uzbekistan’s arguments were thus not dismissed simply by virtue of the narrative of the Kim investors. 508 They were rejected because the Kim investor narrative was contextualized in the context of Uzbekistan town laws and actions.509 Finally, the presumption matrix developed on the basis of the principle that good faith should not be blind to the interests of third parties in the outcome of the litigation. Again, the relevant estoppel analysis would support the view that reasonable reliance by third parties is sufficient to create an estoppel.510 This should apply by analogy in the context of presumptions. The factual findings made by international courts and tribunals have significant downstream effects by creating a foundation for future legal relations between non-parties. They thus affect more than the parties. The use of presumptions should not be blind to this concern. Again, the Corfu Channel case demonstrates that this concern is already implicit in the use of presumptions when presumptions are used well. Thus, the impact of the judgment of the Court on tug and barges on October 29th, 1946, as established by the affidavit of Enrico Bargellini, which was not seriously contested.”). 504. See id. (“As the Parties agree that the minefield had been recently laid, it must be concluded that the operation was carried out during the period of close watch by the Albanian authorities in this sector. This conclusion renders the Albanian Government's assertion of ignorance a priori somewhat improbable.”). 505. Kim v. Republic of Uzb., ICSID Case No. ARB/13/6, Decision on Jurisdiction, ¶¶ 546-87 (Mar. 8, 2017). 506. Id. at ¶ 546. 507. Id. at ¶ 587. 508. Id. 509. Id. 510. FARNSWORTH, supra note 475, at 100-01. 2020] TRUTHS IN TRANSLATION 183

Yugoslavia was one of the motivating factors in the use of presumptions.511 Yugoslavia was not a party to the proceedings. The Court therefore had to tread carefully in making factual findings that would negatively impact Yugoslavia’s rights and interests. The use of presumptions allowed the Court to do this by avoiding the question who mined the straights. The change in perspective will have a marked impact on other cases. The Article began with a discussion of the Pulp Mills case between Argentina and Uruguay.512 That case may well have been decided differently in the context of the translative approach to presumptions. As discussed above, the Court in Pulp Mills decided that there was no evidence on the basis of which a violation of the underlying environmental obligations incurred by Uruguay for building pulp mills on the banks of the river could have been established.513 But the Pulp Mills Court also found that Uruguay failed to live up to its procedural obligations to consult with regard to new projects on the river.514 This failure to communicate is itself a tell. It makes it reasonable for Argentina to conclude that it would have had grounds to object had it been meaningfully consulted. Thus, here we have a case in which Argentina was deprived of the procedural opportunity to communicate and consult, and then lost the ensuing litigation on the basis of Uruguay’s narrative. The matrix developed in this Article makes this result reasonably suspect. It makes it more supect considering the impact of the Court’s conclusions will be felt by people, flora and fauna on the river—parties who were intended beneficiaries of the underlying treaty, but not parties before the Court. A change in our approach to presumptions thus is not a distinction without a difference but a means to make international justice significantly more responsive to the expectations of the global community in the rule of law. Even as the deployment of polyvalent presumption can change the outcome of cases, it is important to note what they do not do. We still are no closer to finding out what actually happened. The use of presumptions does not avoid the need to construct reality in order to do material justice. Or differently put, historians studying archival and archeological evidence in the distant future

511. Corfu Channel, 1949 I.C.J. at 16-17. 512. Pulp Mills, 2010 I.C.J. at 15. 513. Id. at 106. 514. Id. at 70. 184 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 may very well come to radically different conclusions as to what actually occurred than the adjudicative processes. Again, the Corfu Channel case is telling. Recent archival evidence together with archeological evidence suggests that the British entered the straights with an intent to provoke a skirmish with Albanian forces.515 Albania had apparently anticipated such a potential move by the British on the basis of the unsanctioned removal of mines by British warships from Albanian waters.516 In response, Albania communicated with Yugoslavia to seek out a common defense policy for the area premised upon Yugoslav mining the straights.517 A communique sent by Albania to Yugoslavia after the British vessels hit mines stated that “[t]he warship that hit a mine within the zone mined by us bore the number RE 62.”518 In other words, the International Court of Justice was historically wrong on key factual questions before it which it resolved on the basis of presumptions: British passage was not innocent,519 and Albania had, in fact, colluded with Yugoslavia.520 The translative use of presumptions thus has a different goal from finding out historical fact. Such a search would fail to resolve sensitive disputes when they occur (i.e., it does us little good today to resolve the Corfu Channel dispute by pacific means). The point rather is that the ultimate decision should be made in a manner that creates a means for engagement and dispute resolution between the narratives created by the parties. The rule of law does not impose material justice from an external, objective, fixed source. It anchors material justice in the expectations of participants in transnational legal processes. It does not create a narrative of its own. It provides a bridge for us to understand the narratives of others and making decisions from this new, communicative vantage point.

515. See Ana Lalaj, Burning Secrets of the Corfu Channel Incident 14 (Woodrow Wilson Int’l Ctr. for Scholars, Working Paper No. 70, 2014), https://www.wilsoncenter.org/sites/default/files/cwihp_wp_70_burning_secrets_of_the _corfu_channel_incident.pdf [https://perma.cc/65U5-ZLXY]. 516. Id. at 8, 19-21. 517. Id. at 19-21. 518. Id. at 21 (emphasis added). 519. Id. at 14. 520. Id. at 21. 2020] TRUTHS IN TRANSLATION 185

VI. CONCLUSION This Article highlighted the important role of presumptions in international dispute resolution. It critiqued that the current doctrinal use of presumptions runs into an important is-ought problem: it assumes that things are as they ought to be. Moreover, it borrows how things ought to be from the self-regulatory horizons of the actors whose conduct is at issue in disputes at bar. This creates significant problems for fact finding in international dispute resolution as it premises fact finding in narratives and normative claims rather than empirical fact. It is subject to the charge of peddling in alternative facts. The Article suggested that this problem might well be overcome if the focus of presumptions is not on the actor’s conduct, but rather focused on the relationship between actor and the community acted upon. It suggested that this perspective could be anchored in the principle of good faith as other regard. This good faith as other regard could translate the claims brought forward by actors and acted upon into the normative framework of the respective other. It thus presented a bridge through which fact finding could always be anchored in the horizon of all affected parties by dispute resolution. This focus upon presumptions anchored in effects of action as communication allows us to make a broader observation. This broader observation is that the mechanism of fact finding through other regard allows us to deal with narrative without need for a tie breaker in an objectively testable, empirical world. It thus allows us to deal with a discourse that is increasingly premised in battling alternative facts and claims to value or respect. Rather than proving one side right by reference to the world, it flips the script and anchors decision and engagement in the principles and values of the respective “other” party opponent. This focus is anchored in, and has been tested by, strategies of alternative dispute resolution. The key benefit of such a strategy is that international dispute resolution—and fact finding in international dispute resolution— can again fulfil the role of pacific dispute resolution. It can serve as a means both to diffuse and to translate claims made in the idiom of alternative facts. It can understand these alternative facts on the basis of the actions that brought them about and the effective assertion of alternative facts imprinted in the world. By translating these alternatives facts, and anchoring decision in a language that 186 FORDHAM INTERNATIONAL LAW JOURNAL [Vol. 44:1 is sensitive to its underlying claims to dignity, dispute resolution will be respectful to the underlying claims to equal regard and equal value that fuels recourse to alternative facts in the first place.521 Thus, it again can serve as the means of de-escalation that its original authors in the Hague in 1899 had hoped it could become.522

521. See FUKUYAMA, supra note 332, at 140-44. 522. DOUGLAS JOHNSON, THE HISTORICAL FOUNDATIONS OF WORLD ORDER: THE TOWER AND THE ARENA 639 (2008).